Alright, thank you for the clarification. Basically my rant was the result of my frustration through all this time, desperately waiting for those materials to come out. Every project I worked on in the past year and half, would have had enormous benefits from the Megascans textures, and I was constantly preaching out the quality of it to my colleagues, waiting month after month, trusting the "coming soon" words on Quixel website, and keep being frustrated over and over because of the many delays.
I want to trust your word on April 30. Please, don't disappoint again.
Thanks Rikk for letting me know. Believe me, I share your frustration and hunger to release already. I get the strong feelings vented in this thread, these thoughts consume me 24/7. It's a constant tug of war, when dedicated to Megascans, which requires much traveling (we are curating scan data on 4 continents at all times), tool updates and general visibility slow down causing frustration all-round, and vice versa. It's a pickle that satisfies no-one until there are results. Until then words are just words, no matter how much work goes into it behind the scenes.
I would like to say also a word to you teddy, i think to wait for something amazing like megascans is ok even when it was let's say "funny" , "frustrating" or what ever, to see soon, yeah soon etc....
What i have read through the internet and experienced by my self is that you guys have just forget about your quixel suite customers and users. And this is really disappointing, in my opinion there is no excuse to cancel the update support and communication for such a long time.
So things like no upadte support for your software, the big deelay of megascans, wrong price release of megascans, canceling the facebook comments, no communication with your customers. This are pretty heavy facts which i never have experienced before. What i basically like to say with that is, you should not be supprised about the heat that you or let's say quixel has experienced. And hopefully the communication will be from now on much more better and also the update support for the quixel suite.
David i understand your frustration and a lot of what you say is true and valid, these subjects have now been brought up, discussed, and answered several times already. Quixel do care a lot about our customers and as described before, it has been an issue of time or rather lack there off that has created the unfortunate situation leading up til now.
As i said it has been asked and answered many times before so maybe we can move on from this now.
If you have any further questions or issues you would like to bring up feel free to email me at eric.ramberg@hmail.com
I am not. If you are so rich, why don't you buy a subscription for me as well, yes? Thank you very much.
$1,000 for a commercial license to textures of this kind of quality would be a really good deal! (We are talking thousands of textures!) You do realise that most commercial software goes well beyond these kind of prices? There are many cheaper texture options out there, but absolutely none that have this kind of quality and detail. I've been saving toward Megascans for quite a while as I feel it will be an extremely worthwhile purchase.
Anyways, not trying to start arguments, excited for April 30, and I'm sure we will all be satisfied on release!
Hey Eric, actually this was not a complain, i'm sorry if there is any missunderstanding in my text. All i wanted to say is, hopefully from now on we will get the support, the updates and the communication we deserve.
David no worries I understand, we're doing everything we can right now to get better and give all you customers the best possible support and it feels like we started to move in the right direction! hopefully the change will be noticable from now on!
Thanks Rikk for letting me know. Believe me, I share your frustration and hunger to release already. I get the strong feelings vented in this thread, these thoughts consume me 24/7. It's a constant tug of war, when dedicated to Megascans, which requires much traveling (we are curating scan data on 4 continents at all times), tool updates and general visibility slow down causing frustration all-round, and vice versa. It's a pickle that satisfies no-one until there are results. Until then words are just words, no matter how much work goes into it behind the scenes.
IMO as a programmer like you, and if I mention right you're doing it all alone, there is no space to say: "it's has to be released since two years and wtf why it takes so much time?" I believe, to be honest, if we could see the source code and algorithms we would say damn I'm sorry Teddy and the whole Quixel crew.. Damn every map is scanned from all around the world, 100% PBR, look at the MEGASCANS shots some earlier pages here- take the time you need to finish the Suite and especially MEGASCANS ! There is happening enough.. Hello? We're can talk directly with the devs and can get support from them since hours, that's great and a nice manner.
Thanks for reading and keep in mind that you haven't paid for the final version but for the actually great 1.8 and because of that it doesn't make sense to complain about ^^
The only thing that rubs me the wrong way about Megascans, and I hope this makes sense, is the fact that the library has literally thousands of scans (about 7000 according to the GDC post).
For such a small company one would think it'd be easier to start with much lower numbers. People are getting the impression this is getting forever to be released but I bet most people aren't expecting Megascans to be as insanely huge as it's going to be when it launches.
I could give you a number, but it would be based solely on speculation. I would rather base it on facts. But for funny sake, lets go with 500TB of storage and that much bandwidth per month. Using Amazon would run : $47,769.29/month. (if my math is correct. I can't think straight right now).
Basically if you could pass some sort of idea of how much storage space/bandwidth we are talking about?
I could give you a number, but it would be based solely on speculation. I would rather base it on facts. But for funny sake, lets go with 500TB of storage and that much bandwidth per month. Using Amazon would run : $47,769.29/month. (if my math is correct. I can't think straight right now).
Basically if you could pass some sort of idea of how much storage space/bandwidth we are talking about?
So then, about 100$ a month for each Terabyte? Sounds reasonable and provides zero excuses as to 'Bandwidth" or where Bandwidth affects pricing.
Please, lets not restart this discussion again based on loose assumptions. Quixel will not use bandwith costs as an excuse to charge more, it is however a large factor due to the materials taking up incredible amounts of space.
Please, lets not restart this discussion again based on loose assumptions. Quixel will not use bandwith costs as an excuse to charge more, it is however a large factor due to the materials taking up incredible amounts of space.
Well that's why I asked for more information. Did quixel not ask for advice on pricing? If your going to be extremely vague, then no matter what you go with is going to piss off people. Why are you being so secretive about how much storage we are talking about?
Vic my comment was not pointed at you at all! Here is some more information about Megascans:
It depends a little on the resolution: for an 8K material with 8bit maps and 32 bit displacement/bump it's around 350MB per material. An object scan with 8K maps (+32 bit displacement) and ztool, highpoly and LODs is 700MB.
Around 80MB per material for 4K, and 200MB for objects with 4K materials and source highpoly meshes and ztools.
At the moment the library holds around 3,000 scanned assets (tileable surfaces, vegetation atlases, natural objects and scatter object packs), and a few thousand scanned brushes. Not everything will all be available at launch but added on little by little during the beta as testing progresses.
All in all the entire library (excluding brushes) takes up little over 1TB at max resolution @ 8 bit and 32 bit displacements.
Well that's why I asked for more information. Did quixel not ask for advice on pricing? If your going to be extremely vague, then no matter what you go with is going to piss off people. Why are you being so secretive about how much storage we are talking about?
because it very quickly gets the discussion off topic, as users start comparing the cost of thier own web hosting without a measure of what Teddy needs Megascans hosting to do (super sized, super quick super reliable and ecommerce tied into one).
Quixel are asking specifically:
"How much do you think it's worth to YOU the user to be able to access this content."
Not:
"How much do you think it cost us to generate this content and upload it to the web, and make all the hard work that went into it pay off financially for us."
The backend is irrelevent (to us) its purely a notion of putting a price on the value of megascans to your work.
As an example, for a 2K 'charred wood' scan i'd be thinking somewhere in the $00.50 range, and would probably pay $1.00 for a 4K. I would presume i'd could be buying a lot of scans, so would be most attractive to me for individuals to be in line with Itunes prices. Would that feel too cheap or expensive to you?
And I think that sadly, as indicated by the level of upset i've seen on this thread Teddy & Co are already pissing people off left right and centre, though I can't imagine why.
It's not like any of the games developers in the industry have much in the way of transparancey iwht their user base (unless it's Peter Molyneux speculating about features he's invented mid sentence, or a kickestarter (which is basically the devs form of a investor report, rather than PR)).
If Nintendo can get away with holding their cards close to their chest, cut the guy busting his ass over in Sweden some slack I say.
Vic my comment was not pointed at you at all! Here is some more information about Megascans:
It depends a little on the resolution: for an 8K material with 8bit maps and 32 bit displacement/bump it's around 350MB per material. An object scan with 8K maps (+32 bit displacement) and ztool, highpoly and LODs is 700MB.
Around 80MB per material for 4K, and 200MB for objects with 4K materials and source highpoly meshes and ztools.
At the moment the library holds around 3,000 scanned assets (tileable surfaces, vegetation atlases, natural objects and scatter object packs), and a few thousand scanned brushes. Not everything will all be available at launch but added on little by little during the beta as testing progresses.
All in all the entire library (excluding brushes) takes up little over 1TB at max resolution @ 8 bit and 32 bit displacements.
Let me know if you have anymore questions!
Ah ok then. Thanks for the info, that gives me, at least, something to make an educated guess off of. Thanks I will review and make a recommendation based off my needs and what would appear to be your needs.
It looks like with that info, someone like me who would only use the 2k images and probably anywhere from 1-10 scans a week, I would cost you about $2-$12/month in data and BW.
So what I would do, is break down in 2k, 4k, and 8k access groups. Say start at $19/month.
2k group limited to 20 scans a month. Giving options to add download limits to each group. So if I wanted to add another 20 scans a month, it would be $10 more. Go up from there. (you would have to do your own math for the rest), but I would add $10/month for each increment. Have a Commercial unlimited plan that is in the $1000's
I feel as long as you have a plan on the lower end that's not like one scan a month, and is reasonable like the $19/month I suggested, I would not be outraged by that. (being mostly hobby and all) I am working on a commercial project at the moment, so they would be useful in my level designs.
TL:DR Start small (remember the little guy) and grow it from there.
Vic my comment was not pointed at you at all! Here is some more information about Megascans:
It depends a little on the resolution: for an 8K material with 8bit maps and 32 bit displacement/bump it's around 350MB per material. An object scan with 8K maps (+32 bit displacement) and ztool, highpoly and LODs is 700MB.
Around 80MB per material for 4K, and 200MB for objects with 4K materials and source highpoly meshes and ztools.
At the moment the library holds around 3,000 scanned assets (tileable surfaces, vegetation atlases, natural objects and scatter object packs), and a few thousand scanned brushes. Not everything will all be available at launch but added on little by little during the beta as testing progresses.
All in all the entire library (excluding brushes) takes up little over 1TB at max resolution @ 8 bit and 32 bit displacements.
Let me know if you have anymore questions!
1TB full only with megascans materials/ meshes, I'm in love more than before Christmas time this year: April, 30th :poly145:
I was talking to Teddy and asked him if there was anything we could show to the people of Polycount as a mini-update. So he suggested we showed one of the mini-environment packs (probably not the final name). There will be full sized environment packs with more stuff, but these small ones are kinda nifty, they are very useful. They containt the following:
So a total of 20 assets. And this pack will be sold for a price of $49 at 2K resolution to hobbyists that DON`T have a subscription. So this is a one time purchase prize. And here is a scene from Cryengine using all the included assets:
It looks really fantastic, no question about that. But the price is a waaaaaay to high, it's almost the same price orientation like on the UE4 Marketplace. So at least for me at this price constilation megascans is simply to expeniv and we talking just about 2K textures...
I also can't imagine that idies will pay that much even when it looks really great.
Any news about a new quixel suite version ? Can we also count on the 1.9 release of quixel suite when megascans start?
$50 for 20 assets in a pack isn't expensive. But it's probably still too early to talk about prices. We'll have to see how subscriptions / single scans are priced. Pretty keen for the 30th.
Especially when the biggest part of the 20 "asstes" is almost similar. It looks amazing like a wrote before but a realstic price is more like 20 USD. I mean we are talking about 2 x 2K textures, which i guess are the same material and a few small rocks. So i even don't liketoi imagine what the price of tree set is...
@DarkWEdge this is what i also like to know, i'm very curious if the Megascans beta will also bring the long overdued 1.9 release.
One says it's expensive, the other say's it's not... What does this mean?
It simply means it's debatable and you can't please them all... We agree the scans look amazing, so that's something... "shrug"
Lets weigh the do it yourself method in finances, technique, and time... roughly
Sorry Shanteez your example makes abbsolutly no sense to me, no offense to you.
Even when buying equipment for producing your own materials, you can't compare the posibility to make basically a unlimited amount of materias with one purchased pack for 50 USD.
How ever, of course they are people who can spend 50 USD for such a "small" pack. At least i can't afford this amount... I already had a talk with a few guys about the price and the pack, they all say the same. Amazing quality but to expensiv and like i wrote before , i don't think indies have the money to buy this kind of packs. No offense to quixel but they know very well that there is nothing like megascans on the market "now". So they can capitalise they work until more librarys / similar products appear on the market.
Sooner or later more companys will offer this kind of textures and maybe than surfacescans become more affordable for the average Freelancer, Indie or Hobbyst.
I can't imagine that this price strategy is the right way, the whole software "even when not working " has a price of 99 USD and when it finally starts working it can be a huge win for developers. On the other site two packs have basically the same price like the whole software pack, this makes at least for me absolutly no sense.
About "or the alternative paying 50 bucks for ultra high resolution scans" we are talking about 2K, where exactly did you see the ultra high resolution?
One says it's expensive, the other say's it's not... What does this mean?
It simply means it's debatable and you can't please them all... We agree the scans look amazing, so that's something... "shrug"
Lets weigh the do it yourself method in finances, technique, and time... roughly
or the alternative paying 50 bucks for ultra high resolution scans
I don't think you'll get nearly the same results doing it yourself, unless you build your own scanner (here is a video of the setup Quixel have: [ame]https://www.youtube.com/watch?v=8alYZgkwClM[/ame]). Also think about all the work needed to generate the various map types, ensuring the maps stay consistent (accurate), and ensuring they all tile correctly.
Also these packs seem to be aimed more at people who just want to quickly dress up a scene/ have some good looking and similarly themed assets to work with. The real fun will be the actual online library with 8K scans/models. 10 more days, counting them down
You are absolutly right, probably you wont get the same quality until you have real experience in this kind of "let's call it science :-)".
Of course there is a lot of work but they are simply to expensiv at least in this kind of packs.
And for the subscription, what are you expecting? When such a small pack is priced at 50USD you won't get a subscription for like 29 USD and be able to download the same content.
The Unreal marketplace offers 20 x 2K scannded surfaces textures, for only 29 USD!
Maybe the textures are not 100% megascans quality but honestly can you tell a dofference ? especailly later in the engine ? I can't...
You are absolutly right, probably you wont get the same quality until you have real experience in this kind of "let's call it science :-)".
Of course there is a lot of work but they are simply to expensiv at least in this kind of packs.
And for the subscription, what are you expecting? When such a small pack is priced at 50USD you won't get a subscription for like 29 USD and be able to download the same content.
The Unreal marketplace offers 20 x 2K scannded surfaces textures, for only 29 USD!
Maybe the textures are not 100% megascans quality but honestly can you tell a dofference ? especailly later in the engine ? I can't...
Yes, I imagine there was thousands of hours of trial and error and understanding needed to produce the content for Megascans.
I do think that for example a 3 month Megascans subscription with a daily quota (or whatever the limit is), while definitely would be more expensive, would easily get you way more content than is in the packs- and in a much higher resolution, etc- we will have to wait and see.
As for those marketplace materials- they actually look pretty decent- I guess people can choose to use whatever they want! There are always alternatives- I come from a music background and you always have to pay a premium for the absolute best source material- drum hits recorded in pristine 24 bit quality in a state of the art studio vs drum hits some guy recorded in his room for a fraction of the price- the quality/price rule doesn't always apply, but in many cases the best is expensive.
I absolutly agree, of course quality has his price "usually". But sometimes companys overprice they're products in such a heavy way, that it starts to become absurd.
I'm simply happy that there is "in this early phase of PBR" alreday competition or let's say a alternative for getting good looking but affordable surface textures.
I'm also very curious about the other offers, only 10 days left and we will see where the tour goes... But the announcing of the 50 USD pack is alreday the second price shock after the "unintentuional" market place release of quixel for 150USD
Have not seen the actual assets, only demo scenes. I have a feeling the megasscan library will be top notch stuff though.
Check this video out showing the process Epic used for the Kite demo photoscanning some 250 assets.
They traveled to NZ to do it and i can imagine it cost them quite a lot. I do a lot of photoscanning and getting rid of the AO and making good maps is pretty hard and very time consuming. I have camera , light equipment costing more than 10 000 USD. I an still very much interested in megascans. I also do my own HDRi panorama light maps. I finance my projects mostly out of my own pocket though so i hope the price will be reasonable.
BTW anyone know how Epic cancels out the AO in the linked video? Have been using tone mapped HDRi images with some success but the technique they use with a gray ball and mirro ball seems interesting. Don´t know how its actually done though.
I like the pricing model of GameTexures.com where there are different monthly subscription pricing tiers based on the maximum number of textures that can be downloaded per month from the library. This nicely solves the data bandwidth and storage costs since the subscription price varies based on the maximum data that can be downloaded per month. It also provides a natural tiered price structure that can easily accomodate both indies and large commercial studios. Finally, it provides a continuous source of sustained revenue to fund an ever growing texture library.
$19-$20 / month seems to be a price point that works for most indies for most third party game development products (SpeedTree, Maya LT, etc).
GameTextures.com offers:
$19/mo for 75 2k PBR texture downloads per month
$30/mo for 200 2k PBR texture downloads per month + 2 texture requests
$1500/yr ($125/mo) for unlimited downloads per month and multiple seats + 10 texture requests
I think that an additional pricing feature that would help significantly to keep customers motivated to maintain their subscription would be to use a subscription model similar to TMobile that allows customers to keep their unused minutes. That is, the monthly texture downloads would be cumulative, with unused downloads available on subsequent months. This means that customers don't feel like their subscription money is being wasted if they are not downloading lots of textures every month because they are tied up doing other work, so the customers are much more likely to stay subscribed.
The problem with asset packs for a single type of texture category such as forests or mountains when buying textures from a texture library is that the indie customer may not need all the textures in the pack and may need other textures not in the pack. For indies on a limited budget, this can easily put the cost out of reach since they need to buy many different packs to get a broad enough texture selection for a competitive game and don't necessarily need all the different assets in each pack.
Commercial studios don't have a problem since they are always going to buy an unlimited commercial license anyway, since texture library costs are a very small part of their overall development budget, and they can't afford any wasted time or limitations of any sort.
However, as Unity3d and UE4 have shown, the 3d development market is changing considerably and is quickly moving to very large numbers of independent developers with small budgets developing for many different platforms including smart phones, and not just a few large studios with large budgets developing for consoles. Unity3d also demonstrated that there is a huge market potential in that large indie development market. What it lacks in high priced margins it more than makes up for in very high volume.
For the indie market, texture asset packs based on a fixed number of texture downloads that can be selected from the entire library makes much more sense. That way, the indie developer can strategically select only the assets they need from the library and balance the number of textures in each category with the number of texture categories for their particular project.
The problem with texture asset packs from the vendor point of view is that they represent non-recurring revenue, they are a one-shot purchase. Therefore, it's in their interest to motivate customers to buy a library subscription. This works for the indie customer if it is cheaper over time than buying texture-selectable asset packs.
What I see as an ideal texture library pricing strategy would therefore look something like the following:
1. Various sized texture asset packs that allow the customer to download a fixed number of customer-selectable textures from the entire library for a fixed price. There would be several different sized packs with the larger packs cheaper per-texture than the smaller packs.
2. A set of library subscription tiers (say 3 tiers, an indie tier, a freelance tier, and a commercial tier). The only difference between the tiers is the subscription price and the number of texture downloads per month allowed (and maybe the right to request a certain number of customer-defined textures per year that end up available to everyone in the library). The downloads in the 2 lower tiers are cumulative across months so unused downloads are available in subsequent months and are never lost (so it ends up being a way to gradually buy textures from the library over time at a given price per texture). The highest commercial tier is a yearly subscription and allows unlimited downloads (it is yearly to guarantee a certain up-front payment to allow for an unlimited download capability). A library subscription allows the customer to select and download any textures from the library up to that month's download limit. Library subscriptions are all cheaper per-texture than texture asset packs since they provide recurring revenue to the texture vendor. Also, the higher tier subscriptions are cheaper per-texture than the lower tier subscriptions.
Both the asset packs and the lower tier subscriptions would be priced to allow indies to buy enough customer-selectable texture assets over time to create very richly textured games on a realistic indie budget (say $20/mo for the texture assets), with the subscriptions costing less per texture than the packs. As mentioned above, commercial studios won't bother with the limitations of asset packs and lower tier subscriptions since the limitations are not worth their time, so the packs and lower tiers can be priced to provide indie and freelance customers a very broad set of textures over time at a price that meets their limited budget.
Since the cost of producing textures is a sunk cost, that has been incurred regardless of how many copies of the textures are actually sold, it pays to spread that cost across as many customers as possible. Therefore the marginal cost of providing textures to a new customer is very small (the download cost), but the marginal revenue earned is much higher, which means that the optimal pricing strategy would be to acquire as many customers as possible, even if it meant lowering the price per copy of the texture.
Great suggestions and i agree. I would much prefer to pay per texture instead of packs if possible.
"For the indie market, texture asset packs based on a fixed number of texture downloads that can be selected from the entire library makes much more sense. That way, the indie developer can strategically select only the assets they need from the library and balance the number of textures in each category with the number of texture categories for their particular project."
An additional pricing strategy that would strongly motivate users to keep their subscriptions over time would be to gradually increase the number of textures that can be downloaded each month for the lower two subscription tiers. For instance, for the indie subscription tier, you could allow x additional textures to be downloaded per month for each month continuously subscribed. For the freelance tier, you could allow 2x additional textures downloaded per month for each month continuously subscribed. The download rates would eventually be capped at the same average cost per-texture as the unlimited tier. That means that regardless of what tier you are subscribed at, you can eventually reach the same average cost per-texture as the unlimited tier if you stay continuously subscribed long enough. If you cancel your subscription and re-subscribe later you are treated as a new subscriber and start at the beginning download rate again. However, if you upgrade your subscription to a higher tier you get to keep your earned additional downloads per month.
This rewards customers who remain subscribed over time rather than subscribing for a short period, downloading a set of textures that are currently needed, then cancel and wait to subscribe again later to download more textures.
That's a great idea for encouraging continuous subscriptions, SA2.
I'd also like to see some type of feature that keeps track of which textures you've downloaded and links those to your license key. I'd hate to download 100+ textures over time, then lose them in a hard drive crash or a RAID array failure and have to slowly build them back up again with a limited number of texture downloads allowed per month. Maybe downloads of previously-downloaded textures don't count against your monthly allowance, or at least count less?
I agree that once you have downloaded a particular texture, that downloading it again later should not count toward your monthly download cap. It's sort of like a smart phone app, once you have downloaded it, it is considered available to use, and if you download it again later it is considered recovering it from a lost or damaged state, not another purchase.
Fraid there's bad news buddy. The Quixel team made a post the other day to let everone know that someone on the team had a bit of unwanted personal mishap/upset. Consequently there's been a delay and nothing will be coming till further notice. I expect they'll issue a new date when they can bang their heads together and figure out what's doable.
I thought it was my job to bum people out and whip up the hate storms, Oskarkeo I´ll let it slid this time!
odigames, on a more serious note I´m very sorry to say that what Oskarkeo is saying is correct, after two unrelated but very serious personal events affected the key members of the Megascans team it was impossible to keep the 30th Deadline.
I´m not sure there will be a new date announced, I think that when its done it will be released, in order to not create another situation like this again.
Touching upon some things discussed in this thread, the content of the packs will be available to subscribing customers as part of the library. There will be different tiers, ranging from Hobby to Commercial license. There would be a limit to how many textures per month you can download, differing depending on what Tier you subscribe to. I will see if the Megascans team feel comfortable with releasing more details regarding pricing and the terms of the subscription.
This might be slightly off-topic, but is there any repository of materials for DDO that people can download? There are similar libraries for Substance and ZBrush users out there : http://pixologic.com/zbrush/downloadcenter/library/
Not yet, no - we're working on that, however! In the future, I would ask that you please create a new thread instead of bumping older ones like this. Cheers!
Replies
I want to trust your word on April 30. Please, don't disappoint again.
What i have read through the internet and experienced by my self is that you guys have just forget about your quixel suite customers and users. And this is really disappointing, in my opinion there is no excuse to cancel the update support and communication for such a long time.
So things like no upadte support for your software, the big deelay of megascans, wrong price release of megascans, canceling the facebook comments, no communication with your customers. This are pretty heavy facts which i never have experienced before. What i basically like to say with that is, you should not be supprised about the heat that you or let's say quixel has experienced. And hopefully the communication will be from now on much more better and also the update support for the quixel suite.
As i said it has been asked and answered many times before so maybe we can move on from this now.
If you have any further questions or issues you would like to bring up feel free to email me at eric.ramberg@hmail.com
$1,000 for a commercial license to textures of this kind of quality would be a really good deal! (We are talking thousands of textures!) You do realise that most commercial software goes well beyond these kind of prices? There are many cheaper texture options out there, but absolutely none that have this kind of quality and detail. I've been saving toward Megascans for quite a while as I feel it will be an extremely worthwhile purchase.
Anyways, not trying to start arguments, excited for April 30, and I'm sure we will all be satisfied on release!
IMO as a programmer like you, and if I mention right you're doing it all alone, there is no space to say: "it's has to be released since two years and wtf why it takes so much time?" I believe, to be honest, if we could see the source code and algorithms we would say damn I'm sorry Teddy and the whole Quixel crew.. Damn every map is scanned from all around the world, 100% PBR, look at the MEGASCANS shots some earlier pages here- take the time you need to finish the Suite and especially MEGASCANS ! There is happening enough.. Hello? We're can talk directly with the devs and can get support from them since hours, that's great and a nice manner.
Thanks for reading and keep in mind that you haven't paid for the final version but for the actually great 1.8 and because of that it doesn't make sense to complain about ^^
cheers, currve
For such a small company one would think it'd be easier to start with much lower numbers. People are getting the impression this is getting forever to be released but I bet most people aren't expecting Megascans to be as insanely huge as it's going to be when it launches.
http://aws.amazon.com/s3/pricing/
I could give you a number, but it would be based solely on speculation. I would rather base it on facts. But for funny sake, lets go with 500TB of storage and that much bandwidth per month. Using Amazon would run : $47,769.29/month. (if my math is correct. I can't think straight right now).
Basically if you could pass some sort of idea of how much storage space/bandwidth we are talking about?
So then, about 100$ a month for each Terabyte? Sounds reasonable and provides zero excuses as to 'Bandwidth" or where Bandwidth affects pricing.
Well that's why I asked for more information. Did quixel not ask for advice on pricing? If your going to be extremely vague, then no matter what you go with is going to piss off people. Why are you being so secretive about how much storage we are talking about?
It depends a little on the resolution: for an 8K material with 8bit maps and 32 bit displacement/bump it's around 350MB per material. An object scan with 8K maps (+32 bit displacement) and ztool, highpoly and LODs is 700MB.
Around 80MB per material for 4K, and 200MB for objects with 4K materials and source highpoly meshes and ztools.
At the moment the library holds around 3,000 scanned assets (tileable surfaces, vegetation atlases, natural objects and scatter object packs), and a few thousand scanned brushes. Not everything will all be available at launch but added on little by little during the beta as testing progresses.
All in all the entire library (excluding brushes) takes up little over 1TB at max resolution @ 8 bit and 32 bit displacements.
Let me know if you have anymore questions!
because it very quickly gets the discussion off topic, as users start comparing the cost of thier own web hosting without a measure of what Teddy needs Megascans hosting to do (super sized, super quick super reliable and ecommerce tied into one).
Quixel are asking specifically:
"How much do you think it's worth to YOU the user to be able to access this content."
Not:
"How much do you think it cost us to generate this content and upload it to the web, and make all the hard work that went into it pay off financially for us."
The backend is irrelevent (to us) its purely a notion of putting a price on the value of megascans to your work.
As an example, for a 2K 'charred wood' scan i'd be thinking somewhere in the $00.50 range, and would probably pay $1.00 for a 4K. I would presume i'd could be buying a lot of scans, so would be most attractive to me for individuals to be in line with Itunes prices. Would that feel too cheap or expensive to you?
And I think that sadly, as indicated by the level of upset i've seen on this thread Teddy & Co are already pissing people off left right and centre, though I can't imagine why.
It's not like any of the games developers in the industry have much in the way of transparancey iwht their user base (unless it's Peter Molyneux speculating about features he's invented mid sentence, or a kickestarter (which is basically the devs form of a investor report, rather than PR)).
If Nintendo can get away with holding their cards close to their chest, cut the guy busting his ass over in Sweden some slack I say.
So what I would do, is break down in 2k, 4k, and 8k access groups. Say start at $19/month.
2k group limited to 20 scans a month. Giving options to add download limits to each group. So if I wanted to add another 20 scans a month, it would be $10 more. Go up from there. (you would have to do your own math for the rest), but I would add $10/month for each increment. Have a Commercial unlimited plan that is in the $1000's
TL:DR Start small (remember the little guy) and grow it from there.
1TB full only with megascans materials/ meshes, I'm in love more than before
I was talking to Teddy and asked him if there was anything we could show to the people of Polycount as a mini-update. So he suggested we showed one of the mini-environment packs (probably not the final name). There will be full sized environment packs with more stuff, but these small ones are kinda nifty, they are very useful. They containt the following:
2 Tilable surfaces
2 3D assets
12 scatter assets
2 foliage props
2 brush scans
So a total of 20 assets. And this pack will be sold for a price of $49 at 2K resolution to hobbyists that DON`T have a subscription. So this is a one time purchase prize. And here is a scene from Cryengine using all the included assets:
Any feedback is of course very welcome!
I also can't imagine that idies will pay that much even when it looks really great.
Any news about a new quixel suite version ? Can we also count on the 1.9 release of quixel suite when megascans start?
@DarkWEdge this is what i also like to know, i'm very curious if the Megascans beta will also bring the long overdued 1.9 release.
It simply means it's debatable and you can't please them all... We agree the scans look amazing, so that's something... "shrug"
Lets weigh the do it yourself method in finances, technique, and time... roughly
"Hardware" CAnon HDR $1-3500 thousand + Software Photoscan $200 + Software Zbrush $800 + Learning photography, Learning Photoscan, Learning Zbrush etc...
or the alternative paying 50 bucks for ultra high resolution scans
Even when buying equipment for producing your own materials, you can't compare the posibility to make basically a unlimited amount of materias with one purchased pack for 50 USD.
How ever, of course they are people who can spend 50 USD for such a "small" pack. At least i can't afford this amount... I already had a talk with a few guys about the price and the pack, they all say the same. Amazing quality but to expensiv and like i wrote before , i don't think indies have the money to buy this kind of packs. No offense to quixel but they know very well that there is nothing like megascans on the market "now". So they can capitalise they work until more librarys / similar products appear on the market.
Sooner or later more companys will offer this kind of textures and maybe than surfacescans become more affordable for the average Freelancer, Indie or Hobbyst.
I can't imagine that this price strategy is the right way, the whole software "even when not working
About "or the alternative paying 50 bucks for ultra high resolution scans" we are talking about 2K, where exactly did you see the ultra high resolution?
I don't think you'll get nearly the same results doing it yourself, unless you build your own scanner (here is a video of the setup Quixel have: [ame]
Also these packs seem to be aimed more at people who just want to quickly dress up a scene/ have some good looking and similarly themed assets to work with. The real fun will be the actual online library with 8K scans/models. 10 more days, counting them down
Of course there is a lot of work but they are simply to expensiv at least in this kind of packs.
And for the subscription, what are you expecting? When such a small pack is priced at 50USD you won't get a subscription for like 29 USD and be able to download the same content.
The Unreal marketplace offers 20 x 2K scannded surfaces textures, for only 29 USD!
Maybe the textures are not 100% megascans quality but honestly can you tell a dofference ? especailly later in the engine ? I can't...
Yes, I imagine there was thousands of hours of trial and error and understanding needed to produce the content for Megascans.
I do think that for example a 3 month Megascans subscription with a daily quota (or whatever the limit is), while definitely would be more expensive, would easily get you way more content than is in the packs- and in a much higher resolution, etc- we will have to wait and see.
As for those marketplace materials- they actually look pretty decent- I guess people can choose to use whatever they want! There are always alternatives- I come from a music background and you always have to pay a premium for the absolute best source material- drum hits recorded in pristine 24 bit quality in a state of the art studio vs drum hits some guy recorded in his room for a fraction of the price- the quality/price rule doesn't always apply, but in many cases the best is expensive.
I'm simply happy that there is "in this early phase of PBR" alreday competition or let's say a alternative for getting good looking but affordable surface textures.
I'm also very curious about the other offers, only 10 days left and we will see where the tour goes... But the announcing of the 50 USD pack is alreday the second price shock after the "unintentuional" market place release of quixel for 150USD
Check this video out showing the process Epic used for the Kite demo photoscanning some 250 assets.
[ame]
They traveled to NZ to do it and i can imagine it cost them quite a lot. I do a lot of photoscanning and getting rid of the AO and making good maps is pretty hard and very time consuming. I have camera , light equipment costing more than 10 000 USD. I an still very much interested in megascans. I also do my own HDRi panorama light maps. I finance my projects mostly out of my own pocket though so i hope the price will be reasonable.
BTW anyone know how Epic cancels out the AO in the linked video? Have been using tone mapped HDRi images with some success but the technique they use with a gray ball and mirro ball seems interesting. Don´t know how its actually done though.
$19-$20 / month seems to be a price point that works for most indies for most third party game development products (SpeedTree, Maya LT, etc).
GameTextures.com offers:
$19/mo for 75 2k PBR texture downloads per month
$30/mo for 200 2k PBR texture downloads per month + 2 texture requests
$1500/yr ($125/mo) for unlimited downloads per month and multiple seats + 10 texture requests
I think that an additional pricing feature that would help significantly to keep customers motivated to maintain their subscription would be to use a subscription model similar to TMobile that allows customers to keep their unused minutes. That is, the monthly texture downloads would be cumulative, with unused downloads available on subsequent months. This means that customers don't feel like their subscription money is being wasted if they are not downloading lots of textures every month because they are tied up doing other work, so the customers are much more likely to stay subscribed.
Commercial studios don't have a problem since they are always going to buy an unlimited commercial license anyway, since texture library costs are a very small part of their overall development budget, and they can't afford any wasted time or limitations of any sort.
However, as Unity3d and UE4 have shown, the 3d development market is changing considerably and is quickly moving to very large numbers of independent developers with small budgets developing for many different platforms including smart phones, and not just a few large studios with large budgets developing for consoles. Unity3d also demonstrated that there is a huge market potential in that large indie development market. What it lacks in high priced margins it more than makes up for in very high volume.
For the indie market, texture asset packs based on a fixed number of texture downloads that can be selected from the entire library makes much more sense. That way, the indie developer can strategically select only the assets they need from the library and balance the number of textures in each category with the number of texture categories for their particular project.
The problem with texture asset packs from the vendor point of view is that they represent non-recurring revenue, they are a one-shot purchase. Therefore, it's in their interest to motivate customers to buy a library subscription. This works for the indie customer if it is cheaper over time than buying texture-selectable asset packs.
What I see as an ideal texture library pricing strategy would therefore look something like the following:
1. Various sized texture asset packs that allow the customer to download a fixed number of customer-selectable textures from the entire library for a fixed price. There would be several different sized packs with the larger packs cheaper per-texture than the smaller packs.
2. A set of library subscription tiers (say 3 tiers, an indie tier, a freelance tier, and a commercial tier). The only difference between the tiers is the subscription price and the number of texture downloads per month allowed (and maybe the right to request a certain number of customer-defined textures per year that end up available to everyone in the library). The downloads in the 2 lower tiers are cumulative across months so unused downloads are available in subsequent months and are never lost (so it ends up being a way to gradually buy textures from the library over time at a given price per texture). The highest commercial tier is a yearly subscription and allows unlimited downloads (it is yearly to guarantee a certain up-front payment to allow for an unlimited download capability). A library subscription allows the customer to select and download any textures from the library up to that month's download limit. Library subscriptions are all cheaper per-texture than texture asset packs since they provide recurring revenue to the texture vendor. Also, the higher tier subscriptions are cheaper per-texture than the lower tier subscriptions.
Both the asset packs and the lower tier subscriptions would be priced to allow indies to buy enough customer-selectable texture assets over time to create very richly textured games on a realistic indie budget (say $20/mo for the texture assets), with the subscriptions costing less per texture than the packs. As mentioned above, commercial studios won't bother with the limitations of asset packs and lower tier subscriptions since the limitations are not worth their time, so the packs and lower tiers can be priced to provide indie and freelance customers a very broad set of textures over time at a price that meets their limited budget.
Since the cost of producing textures is a sunk cost, that has been incurred regardless of how many copies of the textures are actually sold, it pays to spread that cost across as many customers as possible. Therefore the marginal cost of providing textures to a new customer is very small (the download cost), but the marginal revenue earned is much higher, which means that the optimal pricing strategy would be to acquire as many customers as possible, even if it meant lowering the price per copy of the texture.
"For the indie market, texture asset packs based on a fixed number of texture downloads that can be selected from the entire library makes much more sense. That way, the indie developer can strategically select only the assets they need from the library and balance the number of textures in each category with the number of texture categories for their particular project."
This rewards customers who remain subscribed over time rather than subscribing for a short period, downloading a set of textures that are currently needed, then cancel and wait to subscribe again later to download more textures.
I'd also like to see some type of feature that keeps track of which textures you've downloaded and links those to your license key. I'd hate to download 100+ textures over time, then lose them in a hard drive crash or a RAID array failure and have to slowly build them back up again with a limited number of texture downloads allowed per month. Maybe downloads of previously-downloaded textures don't count against your monthly allowance, or at least count less?
hang on in there in the meantime.
Oskar
odigames, on a more serious note I´m very sorry to say that what Oskarkeo is saying is correct, after two unrelated but very serious personal events affected the key members of the Megascans team it was impossible to keep the 30th Deadline.
I´m not sure there will be a new date announced, I think that when its done it will be released, in order to not create another situation like this again.
Touching upon some things discussed in this thread, the content of the packs will be available to subscribing customers as part of the library. There will be different tiers, ranging from Hobby to Commercial license. There would be a limit to how many textures per month you can download, differing depending on what Tier you subscribe to. I will see if the Megascans team feel comfortable with releasing more details regarding pricing and the terms of the subscription.
Vaporise.
ok sad news but hope things go well and when its done its done
good luck and thanks for the hard work already...
odingames - Thank you very much for your kind words!