I'm importing my .objs, normal maps and diffuse maps. For some reason when I'm using DDO for materials and texturing, it looks very pix-elated. I don't know if I made my model too small if my maps are too small... I don't know what the issue is. I don't know how to fix this problem if there is a setting issue taht I'm not aware of. I haven't been able to find a good resource for understanding the settings in Maya or my prospective modeling program into Quixel Suite. If there are certain settings or size issues I'm not aware of, is there anyway you can let me know what those are or how to fix them. I'm trying to figure it out and I can't. I re-imported those parameters several times, I still can't find the solution. If there's a better learning resource please point me to it. If its a subdivision issue, please also inform me of that. If the subdivision for my high poly model isn't high enough, then please tell me.
it's been a while since GDC. anyone back from business trips yet to talk roadmap or something?
@ken0y thats just sketchup geometry and basically vertex color. thats going to be converted into a color id map and a variation is going to become a poor mans height map to be converted via ndo into a normal map (black elements are heavy indentations)
the final low poly mesh is the same just with all the cuts for vertex color striped from the one you saw. (sketchup can do a cleanup like that automatically so no manual work required)
I talked to teddy yesterday for a few minutes, They finally settled down after a really intense week(+) of meetings and travel. He told me of some DDO bugs he manage to find and kill lately, we also discussed what kind of updates to release, as a first step we thought it would be kinda cool to show of some stuff from the Megascans beta, what people have made with it so far, any interest in seeing that?
Cool, that's an interesting workflow, I've never thought to generate my height map from the geometry before (via B&W colour ID), I've always just drawn it by hand in PS. I'll have to give that a go!
Yes Im quite aware of that, let me rephrase: Would you guys like to see some cool Megascans beta results while you wait for other updates?
I think we already know that experienced artist with Megascans can do amazing things. Show us some technical details instead, like mesh topology options, MS UI(in a sense of; how does it work in practice) etc.
any chance to get a date or rough estimate for when new information might be available regarding a time frame for bugfixes aka 1.9 of quixel suite?
@ken0y I did setup my ndo maps in photoshop too but figured that sketchup had the better line drawing toolset and I could basically also draw on the object itself which is less confusing than with hard surface uv shells in PS
P.s.: with Sketchup you can then either export to max and convert to vertex color or you can export from sketchup directly as pdf from a top viewport in parallel mode. The pdf you can then open as line art in photoshop (or Illustrator to do more tweakin). thats not what I ended up using but figured out along the way as another possible solution
Jesus, really no offense but i think people give a s**** about to see another 10000 pictures or clips about megascans, yeah they look awesome, no question about that! But i think the users want finally to work with the software they have bought, bug free or be even able to use it all.
So a release date of a quixel suite version that is finally working, is at least for me more interesting than some pictures and videos. I hope quixel is ready to give some extras or what ever for compensation for the bad support in the last months.
I just read that someone here in the forum sold his license or got a refund or something like that, is that right ? I have talked to a view users also from other forums and we are thinking to do the same if there is no update coming soon and i don't mean the quixel soon, i mean with soon like in the next one or two weeks. So who can we conntact to complain ? Cause this is getting more an more ridiculous.
I think the artistic quality of the current version is already fantastic. But there is a clear unbalance between artistic quality and bug fixing to the point of feeling non functional. As the title "to see, not touch", we can see what can do the software but not able to use it. What makes the things worse is not only that doesn't work but the considerable amount of time spent trying to figure out how to make it work, until you finally realize there are so many bugs in the software that not workaround can solve all of them.
That is why the pursue of new features and higher artistic quality makes little sense and shows the intention of continue the unbalance the next versions. Instead as eye candy feels like "oh, not again".
I have asked in the past if there was an user able to use the current version in a production environment (and not as occasional user). Eric said that numerous people can do it. But I'm curious if somebody could give us a direct feedback here about using the current version in an effective production environment. At least one person.
i'm not a production environment, but personally have found suite pretty stable. there's a big learning curve getting up and running. of it all, but as eric explains elswhere, it mostly comes down to photoshop permissions to write files, disc cache, or locked files. never noticed an issue not related to this. And I appreciate that's not ideal, but to be fair, the issue seems to be getting photoshop to reliably do what suite tells it to.
Speaking as someone who's played around with the autodesk "send to x software" features or goz, briging works some (not all) of the time.
also, i specifically named the threat to look not touch, as quixel never specified gdc releases, and I wanted to manage expectations of the very eager community, that statement wasn't coming from anyone at quixel.
Since Megascans seems to be getting closer and closer. Could someone clarify for me what exactly is the DDO Megascans starter kit that comes with the Suite?
I'm not sure whether this is the collection of materials already included within DDO or an different collection that we will have access to once Megascans is online.
I just watched Interstellar... there was a scene where the team went of fighting giant waves on a planet while time for them moved so rapidly that years went by within earth hours. Then they got back to the ship and were greeted by an aged man that had given up hope that they might ever come back...
...and then it hit me. the quixel community is that guy! I think we just discovered a time anomaly!
I have asked in the past if there was an user able to use the current version in a production environment (and not as occasional user). Eric said that numerous people can do it. But I'm curious if somebody could give us a direct feedback here about using the current version in an effective production environment. At least one person.
Have said this before, Quixel is working great for me. I am very surprised so many can´t get it to work. Old or pirated photoshop version? I use latest CC from cloud subscription. Windows 7ultimate 64 bit with 64 GB RAM.
Initially i had some issues that was resolved installing English language photoshop. If I'm in a production environment or not depends how you define it i guess. I have a very simple workflow that I'm very happy with. Make all my maps in Zbrush and with that it works great. Only bug i have come across is importing a smart material in to another smart material and saving it as a custom material. Sometimes it does not work. Other than that i can work with it for hours with out a single crash. Zbrush crashes more to be honest.
Quixel must have thousands of users. f it did not work for most of them i am pretty sure this forum would be completely bombed.
I would like to see some more mega scan demos! Bring on the awesomeness !
@ Ingemar Lundgren, like in every living situation or lets say drama, the estimated number of unreported cases is for sure much higher. I bought quixel suite on the very first day it was released but after realizing that the software is not bug free and i can't work with it propper, i left using quixel suite for months without complaining about it.
But after months and months my patience is over and here i'am complaining about it. They made enough money to hire more people and give a good support and product back! This is simply stinginess to spend money for new employees!
I'm also on the cloud subscription from adobe! But even when some users are using a pirated version of photoshop "which of course is not ok!" quixel suite should work!
I have asked in the past if there was an user able to use the current version in a production environment (and not as occasional user). Eric said that numerous people can do it. But I'm curious if somebody could give us a direct feedback here about using the current version in an effective production environment. At least one person.
My team is currently using 1.8 for texturing, and its working well. Another team on the project tried to use it and had a lot of hard times doing so, but they found work arounds and continue to use it, after all its just a photoshop file. I mean what dDo is, is a glorified masking program, and intelligently stores settings for later use. Thats it. nothing really radical, the only thing that changes this other than what you can already do with photoshop, is the addition of megascans to be used as a mask or as a straight up color map. Megascans really is the only thing that separates dDo from anyone else which is why it makes sense to me why they are focusing on getting that ready for primetime. Algorithmic users will undoubtedly find megascans useful, and again makes this program something unique. It is buggy, yes, I haven't really had the earth shattering issues that some of you have faced, but I have run into a bug or two. There is always a work around I find though. Some artists I work with just use dDo to create layers and then jump in and use photoshop normally. The thing is for me, dDo on a bad day is still better than me doing a lot of this stuff by hand, especially when it comes to adjusting UV's and only having to rebake the normal map (knald for the win with everything else)
My team is fairly small considering its a AAA studio, and we don't have the resources to hire on an entire substance team to create materials for us (which you absolutely need if you want to use the program with any kind of dead line in mind), so dDo works for us. Do I think the Quixel guys are handling this well? no. But do I think its because they aren't trying or doing their best? no, not at all. Quixel is fairly small team as well, remember that. They are going trough a major hire shift, albeit at the worst time haha, but its something to keep in mind.
I'll post my specs for my machine as well to see if it helps any. I've used the program almost without any error (minus a few bugs here and there that didn't stop my texture making)
Intel Xeon CPU E5 1650 3.2GHZ
64GB RAM
GTX 670 (new card is on the way!)
Photoshop CC 14.1
I`m on Photoshop CC as well and we tested this on multiple pcs and windows 7 as well as 8 and 8.1. Loading presets reshuffles color id definitions in all our test cases. (very simple and 100% possible to reproduce). I`m not sure if this works for someone but so far I haven't heard of any proposed solutions to the issue
Do I think the Quixel guys are handling this well? no. But do I think its because they aren't trying or doing their best? no, not at all. Quixel is fairly small team as well, remember that. They are going trough a major hire shift, albeit at the worst time haha, but its something to keep in mind.
+1 Framedworld
@david1983
"They made enough money to hire more people and give a good support and product back! This is simply stinginess to spend money for new employees!"
I think it's easy to overlook the difference between hiring someone in and hiring the right person in. As a CD once told me ' hire the right one and get the performance of two'. If Teddy is propping up ddo/ndo while project managing megascans, there's not going to be a lot of time left around to train up an assistant/dev, let alone get on here to quantify 'as soon as possible'.
I do think whoever's running the facebook page (which is where i notice all of the hype about upcoming releases, hence this very thread) should chill out from hyping till they're very certain of imminent delivery, or at least able to place a concrete date. but then again, name a single time any of us ever were glad to get an inferior product 'on time'.
we can deadline, but the quality always takes as long as it needs. I expect all contributers to polycount can appreciate this from first hand experience.
@ Oskarkeo Sorry but this makes absolutly no sense to me. If you are not reday with your software "quality always take as long it need" than simply don't sell a unfinished or bug contaminated product! Seeling unfinished products with like no support after the release, is simply just money making on our backs! They made alreday enough money so they have no more preasure, beside the users / community complaining preasure.
Some people simply don't understant that you / we are not in the debit of quixel, they are in debit of what i and we have payed for! Megascans is absolutly fantastic but why they started to work on another products when the "old" one is not finished? And this is very simple to answer, cause they give a S#### about they're users and buyers of the quixel suite, they're motivation is just economic! And megascans can provide that, cause this is a fresh product which has to be released for fresh money!
And for the pictures, videos and announcements... It's the same like in the medieval or roman times, give the mob a little to see, to entertain, to keep them quiet. And yeah for some people it still works today.
that sounds maybe hard but thats simply the true, it's all about money / Capitalism. There is no reason to let your users/buyers "alone" with a unfinished product.
And Eric Ramberg is unfortunately in the position to read and maybe answer some of the topic here and there, which is a really thankless task. And again what i'm writing is not against him it is about the money makers at quixel! And i'm also not against money making but don't spoof your buyers and give them a propper support of your software.
IIRC they did hire/train a new programmer recently. And I don't think saying they're motivated entirely by money is true; Teddy first released the original nDo for free before quixel existed. When dDo2 was released ddo 1 was released for free and owners were upgraded to ddo2 for free. Their licensing system is pretty fair (imo). It sucks that there's some significant bugs for some users and that there isn't a lot of communication atm, but from my understanding Quixel is still a fairly small company. Megascans seems like a very ambitious project; I'm not surprised if it takes priority over ndo/ddo currently. I just hope that after it is released Quixel gets a better ratio of bugfixing to development.
@David1983 show me one piece of software on the internet that does not have a bug report.
or since this is a gaming orientated forum, show me a game over the last few years that hasn't been patched post release.
And before calling them a bunch of greedy capatilists, id take a look at the other software you use to make game art and check out how much those guys are charging, cause it aint $99.00
At least i have never and i mean really never bought a game or software which hasn't receive a update, fix etc.. within like 3 to 4 months. Of course software can be buggy, this is absolutly ecusable but not for this time period.
And don't forget, we still don't have a full workable or bug free version, instead of a working version all we get are promises for Pictures and Videos. I simply feel played like many other users. Quixel should simply say to they're users, yeah sorry we made a big mistake! give us a month to fix or to release a new version and everything would be good. Tell us a fu**** date. But no, all we get are promises, pictures and soon yeah soon... to see, not touch...
And i'm not saying the guys are bad, they have created a amazing piece of software and i think megascans will change or let's say improve the quality of games massiv. But at the end i decided to support this amazing software and bought it instead of downloading a cracked version. When supporting a "small" company with buying they're software i expect in return also a good support from quixel. And thats the main point and problem i have.
I'm compeletey on board with every thing you say there, except I don't want a release date, not unless it can be hit. I think the teases raised expectations unnessarily and its just pissed folks off.
I share your frustration with the silence, especially as polycount is the offical support forum, though in fairness, Eric is here troublehooting with everyone that posts, so it's silence on dev updates, not on customer support.
Yeah i'm also sorry for eric, like a wrote before he has the most thankless task here in this section. And for the release date, of course the date should be realisitc, moving and mvoing the date again and again will make the users even more angry.
And fot the good end, this is not about hate to someone, this is about fairness, companys no matter if big or small should never forget, your success rises or drops with us users!
As for everything GDC 2015! The time leading up to and following GDC has been more than extreme -- let's just say I'm really relieved to be back in full gear again.
Megascans open beta is slated for release April 30. We are preparing content and a few 2D and 3D scan samples along with an official announcement. We really don't want to do any more big announcements without releasing some actual content to mess around with. I would be happy to share some updates in this thread however
We recently wrapped up GDC 2015 where a number of interactive demos were running; we're releasing some of the scans from these demos together with the announcement.
I can finally give you some more updates on what's been going on with Megascans the past few months...
The Megascans closed beta has been running since January, and a lot of love has been put into Megascans during this time as we prepare the service for public launch. Here are some of the beta scans we have been preparing for you (about 1/10 of the current library):
We are also trying to provide as many customizable BRDF calibration options as possible before launch:
All scans are currently undergoing a massive update as we have improved on the 2x2 m tiling pipeline with help from the beta testers.
Here are some of the works that have been created by the beta testers 100% with the Megascans beta:
As for samples, we are preparing a few 8K surface tileables, 4K leaf atlases, and game-ready 3D assets (with 12M high-poly source meshes). As for more free stuff, the beta will include a number of free high-res 2D and 3D scans for download, as well as a free environment pack (some of the scenes above have been created with the assets in this pack).
As for the GPU baker & 3D Painting features, these updates have taken almost over 4 months longer than we had planned for, and we currently can't say how much more time they will take given the remaining challenges ahead. Our original release plan was December, but so many unforeseen issues came up at once. These features have been developed with various beta versions of Unity 5, which has been a pretty rocky experience from update to update causing a lot of issues (our bad for jumping the gun, not Unity's). Really sorry about the delay on these features.
For more on what was shown at GDC, here are the two videos we put together to showcase the latest Megascans progress:
A real-time UE4 scene created with a small portion of the Megascans 3D and 2D jungle library:
Yes, everyone will be able to sign up and start testing I can tell you that it's pretty terrifying to have a date to count down to!
Just as a heads-up, we will not be able to open up the entire library for unrestricted download as the bandwidth costs will be ginormous (tens of thousands of dollars O_O), but will have a number of high-res beta samples available for free for most categories, and a more than fair beta discount for unrestricted access (40% off or so during open beta).
Final price will be revealed in the announcement but what I can say already is that it will be very flexible depending on needs (resolution, bit depth, commercial/non-commercial, which type of environments you are most interested in etc). We want this to be affordable to everyone, whilst trying to not undermine what the scans bring to high-end production and the insane effort that goes into building and evolving the operation. Super tricky I must admit but we hope it will be a pleasing model!
Thanks for the awesome update. Can you go in detail about the subscription model perhaps?
Just interested to know if this will be a classic subscription model ala Autodesk/Adobe where you have access to the service as long as you subscribe or something like Epic's prior GDC were you renewed the subscription in order to get the newest updates/added content but could still have access to the available content at the time if you didn't.
Hey tach, sure thing! I'd actually like to open up this discussion to everyone as we are now closing in on release, and everybody should get the opportunity to have an impact on the service before it is set in stone.
Since everyone has very different needs we are currently supporting the following:
Just a single surface scan or 3D scan (for very specific needs)
Full environment packs including ~50 assets each, such as tileable surfaces, 3D scans, scatter assets, scanned vegetation and zbrushes (for when only needing scans for a certain type of environment)
All-Access subscription (3 months, 6 months, 1 year) to the complete library, with instant access to updates (for the all-you-can-eat gourmet experience)
Studio license (1 year commercial access for all artists, everything that has been downloaded can be used forever on any amount of projects)
Everything you download, you will own forever.
Packs are significantly cheaper than getting many assets individually, and the subscription is the most cost effective by far when needing to download a great deal of assets over an extended period of time. Signing up for 6 months gives greater discounts than 3 months, and the 1 year subscription will mean the most bang for the buck if you are intending on downloading a lot.
Each option is fully scalable in terms of pricing, where 8K+, 32 bit, commercial use constitutes the heaviest tier (fairly pricey but very much worth its price), and 1-2K, 8 bit, non-commercial is the lowest tier (dirt cheap).
I hope this clears up some of the uncertainties, and more importantly that it feels fair and flexible. Actual pricing will be revealed soon enough, but hopefully this is at least something that can spark a discussion if there are options you think are missing
This is exactly what worries me and why I mentioned Epic's model. Let me elaborate. All the subscription options that you mentioned are truly nice but at the same time it encourages people (myself included) to download as much as they possibly can during their subscription times in order to build an offline library for themselves, unnecesarily increasing your bandwidth costs quite a bit. Most of these downloaded assets people wouldn't even probably use ever but still the urge to download more than you'd ever need will be there. Being able to download any assets that were available at the time of subscription even after the subscription is over could potentially prevent that.
tach, I think that's a valid concern, but I'm not sure if your suggestion would fix it. With your idea, what's to stop someone from getting a 1-month subscription, and then having lifetime access to a boatload of content for hardly any money? You could even get another 1-month subscription every once in a while to get access to any newer stuff.
I agree, though, that the servers are probably in for a beating once this opens up.
On another note, in terms of pricing, are there any thoughts on adding in another layer (hooray complication!) to differentiate between hobbyists/indies/studios? For example, you could have the 3/6/12-month options each broken down into hobbyists/indies/studio pricing like the Suite licenses. Or is the plan to let things like the different quality tiers (1k/2k/4k/etc) act in roughly the same way?
tach, I think that's a valid concern, but I'm not sure if your suggestion would fix it. With your idea, what's to stop someone from getting a 1-month subscription, and then having lifetime access to a boatload of content for hardly any money? You could even get another 1-month subscription every once in a while to get access to any newer stuff.
That's the whole point. Nothing stops someone from getting a 1-month subscription and owning the whole library for good in either subscription models.
If you go with the classic subscription model, then as I said nothing prevents people from trying to download the entire library and build their offline one since "Everything you download, you will own forever." Not to mention that bigger server costs also potentially imply higher subscription prices.
And with Epic's model it's the exact same, except that the server stress won't be nearly as much because people know they can simply access those assets online whenever they like even if the subscription has expired. That way there isn't an urge to download everything even if you don't need it.
As for people re-subscribing every once in a while, I personally don't see a problem with that if you run Epic's model. It seemed to work for them pretty well as people usually can't help themselves but re-subscribe often when the updates are truly enticing.
For example, say Quixel has spent the last few years building up a massive library for Megascans which includes 1000+ assets. And that's what will be available on release for everyone and then you get maybe 20 new items every few months.
Wouldn't it seem like a more fitting idea to have Megascans release with half the amount of assets and just have the service build up to those 1000 in the next 6 months? Like having 500 assets on release and then having 80-100 new items added each month. Who can resist resubscribing each month for that?
While I do agree that this would be quite nice, the complex of problems with such a model is equating software with an online asset library, which are inherently two very different things. For instance, bandwidth costs for a piece of software that will be downloaded a few times per year can be assumed nominal, eternal online access to hundreds of thousands of very bandwidth consuming files is a very different (and even incalculable) beast.
The UE4 subscription model (which ended up not really being a subscription in practice due to the subscribe/insta-unsubscribe) was ultimately discontinued after only 1 year; it was simply not the way to go for Epic. The software is now free for non-commercial use but (if I dare speculate) relies instead on attracting a larger user base paired with Marketplace activity to sustain the actual business end of things. I'm just guessing of course but this is how I'd make sense of it
For our current customer base, the SUITE installer bandwidth costs roughly 50,000 USD per year due to the huge amount of traffic this single file generates alone. A single high-resolution scan can consume about as much bandwidth as the SUITE installer. Each individual user having access to the entire library at the point of subscription (terrabytes of data) for all of eternity without a monthly subscription to cover the potential bandwidth costs would necessitate such a large up front subscription fee it would be totally unreasonable to anyone wanting to subscribe. We thus have to find a compromise that enables us to keep the price point fair without putting the entire service at risk.
The optimal solution we found is to have a one time fee that grants you full access for a customizable period of time with eternal rights to everything you download, and some form of bandwidth cap per day so that we can ensure not going bankrupt should a single account login get spread around like wildfire for instance.
Sorry for the long winded and boring (and unsatisfying) reply, I feel the need though to be as transparent about this as possible so we can make this discussion as open as possible
Thanks! Not unsatisfying at all. Plus you just added a very important piece of information too (at least I think it hasn't been mentioned before?). The bit about the bandwidth cap per day. That definitely changes things. Very interested to know what this cap will be.
This is a very good step in the right direction. Thank you for taking the time to let us know what's going on. I know I definitely appreciate the updates to my copy of Suite.
Nice to hear whats going on I'm glad to see things moving along in a nice direction. That studio license sounds great, one for the entire company, sounds awesome!
I don't understand the traffic problems these days.... Take a look at all the filehoster like mega and 10000 others... they are also offering huge traffic for low cost. they charge like what? 10 USD ? a month and they still make huge profits. Traffic in Europe is cheap.
The same for Netflix, Maxdome etc.... They are paying for traffic and licensing and they still make money.
It's just an idea to rethink maybe some points of your system and to reorgenize it. Hopefully saying traffic is sooooo expensive is not a trick to rise the price.
Bandwidth cost for huge files is certainly a real issue, if you don't understand the difference between a tiny company like Quixel and a corporation like Netflix, and what that means due to economies of scale, I really don't know what to tell you. Also, it seems a bit silly to get preemptively upset about this while you (nor I) have not even seen what the pricing model(s) look like. Bandwidth is only one element of the cost of running such a service as well, the content needs to be captured and processed and the service needs to be maintained, all of which is very far from free to do. Then you have the years of work in R&D to even make this possible. Even ignoring all of this we have basic supply and demand, there is literally nobody else offering anything close to the quality of what Megascans will provide.
Sounds like a good time to bring back hard copies of the textures on blu-ray disks and ship those to consumers who purchase the whole package. That would save your bandwidth and open up another avenue of approach.
I for one would not mind purchasing the textures or package and wait a few days for shipping. And if for some reason those were lost or destroyed there is a record of purchase and send me more for a nominal fee.
If the prices of the scans are really high, perhaps even use removable hard drives copy scans to them and ship that with an added cost.
Just throwing out ideas.
PS
If this was even remotely helpful please feel free to pm a free copy of Quixel suite and you would have my gratitude.Good luck to you.
Since I think my original post was misinterpreted I'll rephrase: The entire point is to keep the price point fair. If we were to introduce a model that allows for unlimited downloading of hundreds of thousands of very large files for all eternity, and all this for a one time cost, it would simply not be possible to keep the price point fair. Hence we keep it as low as possible by introducing quota (say 20 assets per day by default) and a customizable subscription time depending on need (3, 6 or 12 month access). Then bandwidth costs will be nominal and will not affect the price at all, which is what everyone wants at the end of the day. Doing BlueRays or the like are thus not needed to keep costs down and would in fact both cost more as well as generally mean a worse user experience. I hope this cleared it up
@ Earth Quake, Netflix etc... was just a example of handling huge traffic... All the small filehosters are not huge companys and they are offering a download from 20 to 100 GB a day for 10 USD. The Cheap Amazon servers are also a good way for large traffic.
But my statement about traffic was no offensse it was just a suggestion, cause traffic was never so cheap than now days! A simple blueray should be bigger in costs than traffic, just by the way.
But however it sounds good to me to keep the price fair. I'm wondering if the beta has strated in January, is there any more content to show ?
I'm curious about the megascans, i have sone a few days ago some textures made by photogrammetry, is megascans the same kind of created texture ? same quality ? better quality ?
About the 20 assets daily quota, would that be per single asset (albedo, ao, normals, gloss, spec, etc...)? Or per full asset, i.e. a set of textures for a muddy terrain counts as 1 asset?
As a currently unemployed self-learner, I really appreciate your efforts to make the pricing fair teddy. I for one can understand the difficulties a small company like yours must face, and I thank you for all your hard work.
Replies
Cool aesthetic style there, is that just your geometry and basic materials with diffuse colour? I'd be really interested to know
Cheers,
K
@ken0y thats just sketchup geometry and basically vertex color. thats going to be converted into a color id map and a variation is going to become a poor mans height map to be converted via ndo into a normal map (black elements are heavy indentations)
the final low poly mesh is the same just with all the cuts for vertex color striped from the one you saw. (sketchup can do a cleanup like that automatically so no manual work required)
Cool, that's an interesting workflow, I've never thought to generate my height map from the geometry before (via B&W colour ID), I've always just drawn it by hand in PS. I'll have to give that a go!
Cheers,
K
@ken0y I did setup my ndo maps in photoshop too but figured that sketchup had the better line drawing toolset and I could basically also draw on the object itself which is less confusing than with hard surface uv shells in PS
P.s.: with Sketchup you can then either export to max and convert to vertex color or you can export from sketchup directly as pdf from a top viewport in parallel mode. The pdf you can then open as line art in photoshop (or Illustrator to do more tweakin). thats not what I ended up using but figured out along the way as another possible solution
So a release date of a quixel suite version that is finally working, is at least for me more interesting than some pictures and videos. I hope quixel is ready to give some extras or what ever for compensation for the bad support in the last months.
I just read that someone here in the forum sold his license or got a refund or something like that, is that right ? I have talked to a view users also from other forums and we are thinking to do the same if there is no update coming soon and i don't mean the quixel soon, i mean with soon like in the next one or two weeks. So who can we conntact to complain ? Cause this is getting more an more ridiculous.
That is why the pursue of new features and higher artistic quality makes little sense and shows the intention of continue the unbalance the next versions. Instead as eye candy feels like "oh, not again".
I have asked in the past if there was an user able to use the current version in a production environment (and not as occasional user). Eric said that numerous people can do it. But I'm curious if somebody could give us a direct feedback here about using the current version in an effective production environment. At least one person.
Speaking as someone who's played around with the autodesk "send to x software" features or goz, briging works some (not all) of the time.
also, i specifically named the threat to look not touch, as quixel never specified gdc releases, and I wanted to manage expectations of the very eager community, that statement wasn't coming from anyone at quixel.
I'm not sure whether this is the collection of materials already included within DDO or an different collection that we will have access to once Megascans is online.
...and then it hit me. the quixel community is that guy! I think we just discovered a time anomaly!
Have said this before, Quixel is working great for me. I am very surprised so many can´t get it to work. Old or pirated photoshop version? I use latest CC from cloud subscription. Windows 7ultimate 64 bit with 64 GB RAM.
Initially i had some issues that was resolved installing English language photoshop. If I'm in a production environment or not depends how you define it i guess. I have a very simple workflow that I'm very happy with. Make all my maps in Zbrush and with that it works great. Only bug i have come across is importing a smart material in to another smart material and saving it as a custom material. Sometimes it does not work. Other than that i can work with it for hours with out a single crash. Zbrush crashes more to be honest.
Quixel must have thousands of users. f it did not work for most of them i am pretty sure this forum would be completely bombed.
I would like to see some more mega scan demos! Bring on the awesomeness !
But after months and months my patience is over and here i'am complaining about it. They made enough money to hire more people and give a good support and product back! This is simply stinginess to spend money for new employees!
I'm also on the cloud subscription from adobe! But even when some users are using a pirated version of photoshop "which of course is not ok!" quixel suite should work!
My team is currently using 1.8 for texturing, and its working well. Another team on the project tried to use it and had a lot of hard times doing so, but they found work arounds and continue to use it, after all its just a photoshop file. I mean what dDo is, is a glorified masking program, and intelligently stores settings for later use. Thats it. nothing really radical, the only thing that changes this other than what you can already do with photoshop, is the addition of megascans to be used as a mask or as a straight up color map. Megascans really is the only thing that separates dDo from anyone else which is why it makes sense to me why they are focusing on getting that ready for primetime. Algorithmic users will undoubtedly find megascans useful, and again makes this program something unique. It is buggy, yes, I haven't really had the earth shattering issues that some of you have faced, but I have run into a bug or two. There is always a work around I find though. Some artists I work with just use dDo to create layers and then jump in and use photoshop normally. The thing is for me, dDo on a bad day is still better than me doing a lot of this stuff by hand, especially when it comes to adjusting UV's and only having to rebake the normal map (knald for the win with everything else)
My team is fairly small considering its a AAA studio, and we don't have the resources to hire on an entire substance team to create materials for us (which you absolutely need if you want to use the program with any kind of dead line in mind), so dDo works for us. Do I think the Quixel guys are handling this well? no. But do I think its because they aren't trying or doing their best? no, not at all. Quixel is fairly small team as well, remember that. They are going trough a major hire shift, albeit at the worst time haha, but its something to keep in mind.
I'll post my specs for my machine as well to see if it helps any. I've used the program almost without any error (minus a few bugs here and there that didn't stop my texture making)
Intel Xeon CPU E5 1650 3.2GHZ
64GB RAM
GTX 670 (new card is on the way!)
Photoshop CC 14.1
Hope that helps anyone.
@david1983
"They made enough money to hire more people and give a good support and product back! This is simply stinginess to spend money for new employees!"
I think it's easy to overlook the difference between hiring someone in and hiring the right person in. As a CD once told me ' hire the right one and get the performance of two'. If Teddy is propping up ddo/ndo while project managing megascans, there's not going to be a lot of time left around to train up an assistant/dev, let alone get on here to quantify 'as soon as possible'.
I do think whoever's running the facebook page (which is where i notice all of the hype about upcoming releases, hence this very thread) should chill out from hyping till they're very certain of imminent delivery, or at least able to place a concrete date. but then again, name a single time any of us ever were glad to get an inferior product 'on time'.
we can deadline, but the quality always takes as long as it needs. I expect all contributers to polycount can appreciate this from first hand experience.
Some people simply don't understant that you / we are not in the debit of quixel, they are in debit of what i and we have payed for! Megascans is absolutly fantastic but why they started to work on another products when the "old" one is not finished? And this is very simple to answer, cause they give a S#### about they're users and buyers of the quixel suite, they're motivation is just economic! And megascans can provide that, cause this is a fresh product which has to be released for fresh money!
And for the pictures, videos and announcements... It's the same like in the medieval or roman times, give the mob a little to see, to entertain, to keep them quiet. And yeah for some people it still works today.
that sounds maybe hard but thats simply the true, it's all about money / Capitalism. There is no reason to let your users/buyers "alone" with a unfinished product.
And Eric Ramberg is unfortunately in the position to read and maybe answer some of the topic here and there, which is a really thankless task. And again what i'm writing is not against him it is about the money makers at quixel! And i'm also not against money making but don't spoof your buyers and give them a propper support of your software.
show me one piece of software on the internet that does not have a bug report.
or since this is a gaming orientated forum, show me a game over the last few years that hasn't been patched post release.
And before calling them a bunch of greedy capatilists, id take a look at the other software you use to make game art and check out how much those guys are charging, cause it aint $99.00
And don't forget, we still don't have a full workable or bug free version, instead of a working version all we get are promises for Pictures and Videos. I simply feel played like many other users. Quixel should simply say to they're users, yeah sorry we made a big mistake! give us a month to fix or to release a new version and everything would be good. Tell us a fu**** date. But no, all we get are promises, pictures and soon yeah soon... to see, not touch...
And i'm not saying the guys are bad, they have created a amazing piece of software and i think megascans will change or let's say improve the quality of games massiv. But at the end i decided to support this amazing software and bought it instead of downloading a cracked version. When supporting a "small" company with buying they're software i expect in return also a good support from quixel. And thats the main point and problem i have.
I'm compeletey on board with every thing you say there, except I don't want a release date, not unless it can be hit. I think the teases raised expectations unnessarily and its just pissed folks off.
I share your frustration with the silence, especially as polycount is the offical support forum, though in fairness, Eric is here troublehooting with everyone that posts, so it's silence on dev updates, not on customer support.
And fot the good end, this is not about hate to someone, this is about fairness, companys no matter if big or small should never forget, your success rises or drops with us users!
You mentioned that Teddy had some fixes already, are they going to be released?
I posted this in the Status & Stability Patch thread, but I'll post it here as well in case any one would miss it
First off, you can grab the latest stability build of the SUITE here: http://www.polycount.com/forum/showthread.php?t=150503
As for everything GDC 2015! The time leading up to and following GDC has been more than extreme -- let's just say I'm really relieved to be back in full gear again.
Megascans open beta is slated for release April 30. We are preparing content and a few 2D and 3D scan samples along with an official announcement. We really don't want to do any more big announcements without releasing some actual content to mess around with. I would be happy to share some updates in this thread however
We recently wrapped up GDC 2015 where a number of interactive demos were running; we're releasing some of the scans from these demos together with the announcement.
I can finally give you some more updates on what's been going on with Megascans the past few months...
The Megascans closed beta has been running since January, and a lot of love has been put into Megascans during this time as we prepare the service for public launch. Here are some of the beta scans we have been preparing for you (about 1/10 of the current library):
We are also trying to provide as many customizable BRDF calibration options as possible before launch:
All scans are currently undergoing a massive update as we have improved on the 2x2 m tiling pipeline with help from the beta testers.
Here are some of the works that have been created by the beta testers 100% with the Megascans beta:
As for samples, we are preparing a few 8K surface tileables, 4K leaf atlases, and game-ready 3D assets (with 12M high-poly source meshes). As for more free stuff, the beta will include a number of free high-res 2D and 3D scans for download, as well as a free environment pack (some of the scenes above have been created with the assets in this pack).
As for the GPU baker & 3D Painting features, these updates have taken almost over 4 months longer than we had planned for, and we currently can't say how much more time they will take given the remaining challenges ahead. Our original release plan was December, but so many unforeseen issues came up at once. These features have been developed with various beta versions of Unity 5, which has been a pretty rocky experience from update to update causing a lot of issues (our bad for jumping the gun, not Unity's). Really sorry about the delay on these features.
For more on what was shown at GDC, here are the two videos we put together to showcase the latest Megascans progress:
A real-time UE4 scene created with a small portion of the Megascans 3D and 2D jungle library:
[ame]
A few of the latest eye candies created with Megascans and the latest version of the suite (now available for download):
https://vimeo.com/121152498
The entire team is steamrolling ahead to bring you something out of the ordinary for April 30th, it's been a long time coming.
- Teddy
It's also nice to have a date to count down to. Can I assume the Open Beta means open to everyone?
Just as a heads-up, we will not be able to open up the entire library for unrestricted download as the bandwidth costs will be ginormous (tens of thousands of dollars O_O), but will have a number of high-res beta samples available for free for most categories, and a more than fair beta discount for unrestricted access (40% off or so during open beta).
Final price will be revealed in the announcement but what I can say already is that it will be very flexible depending on needs (resolution, bit depth, commercial/non-commercial, which type of environments you are most interested in etc). We want this to be affordable to everyone, whilst trying to not undermine what the scans bring to high-end production and the insane effort that goes into building and evolving the operation. Super tricky I must admit but we hope it will be a pleasing model!
- Teddy
Just interested to know if this will be a classic subscription model ala Autodesk/Adobe where you have access to the service as long as you subscribe or something like Epic's prior GDC were you renewed the subscription in order to get the newest updates/added content but could still have access to the available content at the time if you didn't.
Since everyone has very different needs we are currently supporting the following:
Everything you download, you will own forever.
Packs are significantly cheaper than getting many assets individually, and the subscription is the most cost effective by far when needing to download a great deal of assets over an extended period of time. Signing up for 6 months gives greater discounts than 3 months, and the 1 year subscription will mean the most bang for the buck if you are intending on downloading a lot.
Each option is fully scalable in terms of pricing, where 8K+, 32 bit, commercial use constitutes the heaviest tier (fairly pricey but very much worth its price), and 1-2K, 8 bit, non-commercial is the lowest tier (dirt cheap).
I hope this clears up some of the uncertainties, and more importantly that it feels fair and flexible. Actual pricing will be revealed soon enough, but hopefully this is at least something that can spark a discussion if there are options you think are missing
- Teddy
thanks for the update Teddy, great to get the lowdown on whats coming up.
Oskar
This is exactly what worries me and why I mentioned Epic's model. Let me elaborate. All the subscription options that you mentioned are truly nice but at the same time it encourages people (myself included) to download as much as they possibly can during their subscription times in order to build an offline library for themselves, unnecesarily increasing your bandwidth costs quite a bit. Most of these downloaded assets people wouldn't even probably use ever but still the urge to download more than you'd ever need will be there. Being able to download any assets that were available at the time of subscription even after the subscription is over could potentially prevent that.
I agree, though, that the servers are probably in for a beating once this opens up.
On another note, in terms of pricing, are there any thoughts on adding in another layer (hooray complication!) to differentiate between hobbyists/indies/studios? For example, you could have the 3/6/12-month options each broken down into hobbyists/indies/studio pricing like the Suite licenses. Or is the plan to let things like the different quality tiers (1k/2k/4k/etc) act in roughly the same way?
That's the whole point. Nothing stops someone from getting a 1-month subscription and owning the whole library for good in either subscription models.
If you go with the classic subscription model, then as I said nothing prevents people from trying to download the entire library and build their offline one since "Everything you download, you will own forever." Not to mention that bigger server costs also potentially imply higher subscription prices.
And with Epic's model it's the exact same, except that the server stress won't be nearly as much because people know they can simply access those assets online whenever they like even if the subscription has expired. That way there isn't an urge to download everything even if you don't need it.
As for people re-subscribing every once in a while, I personally don't see a problem with that if you run Epic's model. It seemed to work for them pretty well as people usually can't help themselves but re-subscribe often when the updates are truly enticing.
For example, say Quixel has spent the last few years building up a massive library for Megascans which includes 1000+ assets. And that's what will be available on release for everyone and then you get maybe 20 new items every few months.
Wouldn't it seem like a more fitting idea to have Megascans release with half the amount of assets and just have the service build up to those 1000 in the next 6 months? Like having 500 assets on release and then having 80-100 new items added each month. Who can resist resubscribing each month for that?
Time to put on the business hat!
While I do agree that this would be quite nice, the complex of problems with such a model is equating software with an online asset library, which are inherently two very different things. For instance, bandwidth costs for a piece of software that will be downloaded a few times per year can be assumed nominal, eternal online access to hundreds of thousands of very bandwidth consuming files is a very different (and even incalculable) beast.
The UE4 subscription model (which ended up not really being a subscription in practice due to the subscribe/insta-unsubscribe) was ultimately discontinued after only 1 year; it was simply not the way to go for Epic. The software is now free for non-commercial use but (if I dare speculate) relies instead on attracting a larger user base paired with Marketplace activity to sustain the actual business end of things. I'm just guessing of course but this is how I'd make sense of it
For our current customer base, the SUITE installer bandwidth costs roughly 50,000 USD per year due to the huge amount of traffic this single file generates alone. A single high-resolution scan can consume about as much bandwidth as the SUITE installer. Each individual user having access to the entire library at the point of subscription (terrabytes of data) for all of eternity without a monthly subscription to cover the potential bandwidth costs would necessitate such a large up front subscription fee it would be totally unreasonable to anyone wanting to subscribe. We thus have to find a compromise that enables us to keep the price point fair without putting the entire service at risk.
The optimal solution we found is to have a one time fee that grants you full access for a customizable period of time with eternal rights to everything you download, and some form of bandwidth cap per day so that we can ensure not going bankrupt should a single account login get spread around like wildfire for instance.
Sorry for the long winded and boring (and unsatisfying) reply, I feel the need though to be as transparent about this as possible so we can make this discussion as open as possible
- Teddy
The same for Netflix, Maxdome etc.... They are paying for traffic and licensing and they still make money.
It's just an idea to rethink maybe some points of your system and to reorgenize it. Hopefully saying traffic is sooooo expensive is not a trick to rise the price.
I for one would not mind purchasing the textures or package and wait a few days for shipping. And if for some reason those were lost or destroyed there is a record of purchase and send me more for a nominal fee.
If the prices of the scans are really high, perhaps even use removable hard drives copy scans to them and ship that with an added cost.
Just throwing out ideas.
PS
If this was even remotely helpful please feel free to pm a free copy of Quixel suite and you would have my gratitude.Good luck to you.
Since I think my original post was misinterpreted I'll rephrase: The entire point is to keep the price point fair. If we were to introduce a model that allows for unlimited downloading of hundreds of thousands of very large files for all eternity, and all this for a one time cost, it would simply not be possible to keep the price point fair. Hence we keep it as low as possible by introducing quota (say 20 assets per day by default) and a customizable subscription time depending on need (3, 6 or 12 month access). Then bandwidth costs will be nominal and will not affect the price at all, which is what everyone wants at the end of the day. Doing BlueRays or the like are thus not needed to keep costs down and would in fact both cost more as well as generally mean a worse user experience. I hope this cleared it up
- Teddy
But my statement about traffic was no offensse it was just a suggestion, cause traffic was never so cheap than now days! A simple blueray should be bigger in costs than traffic, just by the way.
But however it sounds good to me to keep the price fair. I'm wondering if the beta has strated in January, is there any more content to show ?
I'm curious about the megascans, i have sone a few days ago some textures made by photogrammetry, is megascans the same kind of created texture ? same quality ? better quality ?