Amazon Lumberyard... whats the point?

What cost are you talking about? GameLift is a separate service (if you already use AWS using something like Chef to handle auto deployment is pretty easy and free) and it didn't seem like it was a requirement that it had to be a mult player game so you don't have the AWS cost either.

I agree I don't see what new they offer that will make droves of developers switch. C++ just seems like a bad idea for that kind of tool. At least UE4 has Blueprints that can do pretty much everything you need. I'm sure it will end up being just as popular as Autodesk's Stingray game engine.

What cost are you talking about? GameLift is a separate service (if you already use AWS using something like Chef to handle auto deployment is pretty easy and free) and it didn't seem like it was a requirement that it had to be a mult player game so you don't have the AWS cost either.

I agree I don't see what new they offer that will make droves of developers switch. C++ just seems like a bad idea for that kind of tool. At least UE4 has Blueprints that can do pretty much everything you need. I'm sure it will end up being just as popular as Autodesk's Stingray game engine.

Just from the description on the page it seems this has Something blueprints like : "With Amazon Lumberyard’s visual scripting tool"

And they definately have a market there but not necessarily in the hobby field, it's not because every big guy started worshipping the hobbyist that there are no other markets left, a no risk tool for AAA studio where if you develop a MMO you don't need to manage the hosting, can scale it as your user base does, and your host is the one that designed the engine isn't exactly Something you find everywhere.

Just from the description on the page it seems this has Something blueprints like : "With Amazon Lumberyard’s visual scripting tool"

And they definately have a market there but not necessarily in the hobby field, it's not because every big guy started worshipping the hobbyist that there are no other markets left, a no risk tool for AAA studio where if you develop a MMO you don't need to manage the hosting, can scale it as your user base does, and your host is the one that designed the engine isn't exactly Something you find everywhere.

Oh, must have missed that. Well, makes sense to add that, if CryEngine didn't already bring that functionality (which to my knowledge it didn't).

But such services already exist together with plug and play integrations at least for Unity. Not sure about UE4, but I have a hard time believing this is really an Unique Selling Point for the whole Engine.

Agreed though, these are third party developers mostly. I have no idea about the quality of Amazons services or source code, so it MIGHT be an advantage.

What cost are you talking about? GameLift is a separate service (if you already use AWS using something like Chef to handle auto deployment is pretty easy and free) and it didn't seem like it was a requirement that it had to be a mult player game so you don't have the AWS cost either.

Sure, for a non-multiplayer game the cost might be zero. But I was pointing at the hidden cost of having to DIY if you do not want to go with amazons online service, or the increased cost that comes from this a) being a monopoly and b) the only source of income for the engine, if you DO want to go build an online game.

I know nothing about AWS, so yes, I was quoting secondhand information here. Might be wrong information of course.

Still doesn't sound like an engine to rule them all.

Might be more written for their OWN game development, and put out in the wild to gather some additional bucks from the odd developer that uses their online services... maybe also to just give the engine more credibility, so that people don't go all "what's that? Unknown engine? Must be sh**te!" when they see "build with lumberyard" in amazons own games description.

Though, given the good name it has with gamers because of CryTeks own games, I guess "built with CryEngine" would be the way better sticker to put on the games built with lumberyard....

CryEngine has had visual scripting for a while. Though I've always found the CryEditor unwieldy compared to other editors. I could see something like this being beneficial for say, Armored Warfare. Though that's not really indie, and they went with My.com if I recall correctly for their backend stuff. Twitch integration is kind of neat.

Maybe now we'll get better documentation and tutorials for Lumberyard than we ever got out for CryEngine. That was always my problem with CryEngine.

I honestly don't know how to feel about it. Cryengine, from my experience, has never seemed like a user friendly code base. Even though you don't see the internals of Unity, with a free membership, they at least make adding features like new shaders easier. And Unreal Engine, even with all of it's faults, keeps a clean code base that is extremely easy to use , whether it be in c++ or blueprints. I know that it would of been a large undertaking to create an engine from scratch, but I honestly think that Amazon would of been in a better position than just forking off of Cryengine. I downloaded Lumberyard today and went through some of the code for the renderer and etc, and it still seems to suffer from the same issue I had previously with the engine. Cryengine does not have a clean abstraction layer like Unreal Engine's RHI and Unity's Gfx, so even the rendering of meshes, post processing features, and deferred lighting , require duplicated code for each graphics library. I can probably go on for days about how I feel about Cryengine, but I'll just stop there. The twitch and aws integration is interesting, but I don't think that those features could make up for the shear difficulty that one would have making something of value in this engine, in comparison to Unity and UE4.

UE4: 5% of gross retail sales (more like 10% of post-tax/middleman revenue).
UE4 C++ source access: Free!

Unity Pro: $75/mo per seat, or $1500 per seat.
Unity Free: Free!
Unity C++ source access: Nope. You don't need the source. Ok fine, "Ask us" (aka "If you have to ask, you can't afford it" aka probably $100k+).

Stingray: $240/year per seat.
Stingray C++ source access: "Ask us" (aka "If you have to ask, you can't afford it" aka probably $100k+).

CryEngine: $10/mo per seat.
CryEngine C++ source access: "Ask us" (aka "If you have to ask, you can't afford it" aka probably $100k+).

Lumberjack (aka CryAmazonEngine) : Free. "But please use our (optional) cloud services!"
Lumberjack C++ source access: Free!


Looks good on paper... but knowing people who have worked at CryTek and other people who have made games using CryEngine... they should really be paying you to use their engine. You probably get what you pay for with Lumberjack.

It's also interesting to see that this is how CryTek managed to dodge bankruptcy the other year... They were almost dead, and amazon must've paid them a boatload of money (apparently $50M+) to buy unlimited rights to their source code, which they're now giving away for free. If we assume that a CryEngine license cost $200k, that's equivalent to the income they would've got from 250 AAA games... and now they're pretty unlikely to ever make another engine sale, seeing the Amazon is giving away the farm :lol: So they better spend their windfall wisely. Their days are probably still numbered, seeing that $50M is enough to pay the debts they had and make one more game...

It's also interesting to see that this is how CryTek managed to dodge bankruptcy the other year... They were almost dead, and amazon must've paid them a boatload of money (apparently $50M+) to buy unlimited rights to their source code, which they're now giving away for free. If we assume that a CryEngine license cost $200k, that's equivalent to the income they would've got from 250 AAA games... and now they're pretty unlikely to ever make another engine sale, seeing the Amazon is giving away the farm So they better spend their windfall wisely. Their days are probably still numbered, seeing that $50M is enough to pay the debts they had and make one more game...

I am guessing here, but most probably CryTek wasn't expecting a lot of sales of CryEngine in the next few years... even if 250 AAA Projects would have been ready to fork out for CryEngine in the next 10 years, hell even if 500 would pay the full price in the next 5 years, that is a bad deal when you need the money now.

And given the engine has made as slow sales in the last few years as I think it did (else CryTeks financials would most probably look better), they might have just thought "aw sh*t, this thing will never sell like hotcakes, lets throw it out of the window and hand it over to amazon, and use the money to finance something that sells better".

Maybe they will stop being an engine dev, let amazon take care of that, and concentrate on making games again? Maybe the next Crysis or FarCry is built on Lumberyard?

Not extatic. As a daily Amazon software end-user, notably Amazon Prime Music (... and Fire, which Amazon basically forced upon me even though I didn't want it), I must say that I wouldn't want to touch it with the figurative ten-foot pole.

Even if it's really just CryEngine rebranded, with some stuff added... they sure found a way at Amazon to fuck it up in subtle and not-so-subtle ways. It's annoying but acceptable if Prime Music closes unexpectedly without an error message right after launching, or if it shows the wrong cover and track info (being out of sync with what's selected). Or, if it has duplicate albums of which one is empty, or three dozen other wtf-dont-you-guys-review-what-you-ship kind of bugs. Undeletable content is an annoyance that exists in Kindle, too (which is, apart from being an obnoxious online store employing tactics that are outright illegal outside of online business -- even when you paid extra for the "no specials" version -- for the most part pretty OK, at least it works without major disturbances).

But for a game which, in the end, has my name on it, such stuff isn't just a mere annoyance. It's something I'm not willing to accept if there is any way around it. Not even willing to risk it.

Lumberjack (aka CryAmazonEngine) : Free. "But please use our (optional) cloud services!"

From what I understand (and I didn't look into this in-depth so please correct me if I'm wrong), while technically this might be true the only other option you have is to host your own servers. You are apparently not allowed to use other third party hosting options, so the options are either AWS or self-hosting. Which to me is still not sounding that bad for what you get, but there is some restriction there.

As someone who used EC2 and Amazon's deployment solutions for a while before I got sick of the maintenance aspect and having instances go down only to spin up in incorrect regions I'd be wary of hosting all of my infrastructure with AWS again. But of course this was just one personal project, where I didn't feel like doing all of the ongoing maintenance myself - if you really go all out to set up everything professionally and manage it AWS could be a great option.

I wouldn't be surprised if in a few months amazon launched a relatively cheap games console as a loss leader after roping studios into porting for it. Perhaps a mid range x86 based device with a hdmi port.

They did this with Amazon fire TV stick and the amazon fire TV box, which are marketed as TV and causal gaming devices. With a game engine, some developers and a device they could start offering a games console at a quarter of the price of the xbox one and ps4, using lumberyard as incentive to studios already familiar with cryengine.

Just my own personal thoughts on the possible future...

I wouldn't be surprised if in a few months amazon launched a relatively cheap games console as a loss leader after roping studios into porting for it. Perhaps a mid range x86 based device with a hdmi port.

They did this with Amazon fire TV stick and the amazon fire TV box, which are marketed as TV and causal gaming devices. With a game engine, some developers and a device they could start offering a games console at a quarter of the price of the xbox one and ps4, using lumberyard as incentive to studios already familiar with cryengine.

Just my own personal thoughts on the possible future...

Good point... question is, will a free engine be enough incentive for studios to port enough games to the console, so the game lineup might convince buyers the not-so-great hardware (just guessing here, based on other Amazon devices) is actually worth even the low price asked for it? Its a hen and eggs dilemma with every new console.

The next Ouya?

they sure found a way at Amazon to fuck it up in subtle and not-so-subtle ways

And here is my prejudice confirmed. Against better intuition, I actually downloaded this. Went into a random folder... dev\Code\Launcher\LinuxLauncher and opened a random file (Main.cpp) in an editor. Scrolled down a couple of lines. Right after the function IncreaseResourceMaxLimit you find what?

bool IncreaseSackSizeToMax()

Really, dude. Sack? Granted, you could as well call the function thingamagoop if that's what you wanted, and if people called it with that name it would work fine. But... you know, spellchecking?

What's most amazing is that the function is indeed being called from within RunGame, which means either the guy writing the function used auto-complete and blindly hit return after typing Inc without ever looking what was being inserted (as long as it gives no compiler error, all is good!), or he actually changed Stack into Sack because the compiler did show an error. I'm trying to imagine it was the first one. I'm really trying.

No seriously, I didn't spend hours looking for this... first random file I opened. So with that initial impression, I'm not going to look at another one.

I wouldn't be surprised if in a few months amazon launched a relatively cheap games console as a loss leader after roping studios into porting for it. Perhaps a mid range x86 based device with a hdmi port.

They did this with Amazon fire TV stick and the amazon fire TV box, which are marketed as TV and causal gaming devices. With a game engine, some developers and a device they could start offering a games console at a quarter of the price of the xbox one and ps4, using lumberyard as incentive to studios already familiar with cryengine.

Just my own personal thoughts on the possible future...

Good point... question is, will a free engine be enough incentive for studios to port enough games to the console, so the game lineup might convince buyers the not-so-great hardware (just guessing here, based on other Amazon devices) is actually worth even the low price asked for it? Its a hen and eggs dilemma with every new console.

The next Ouya?

Except porting is probably close to free for all games that currently cryengine so that's a base of AAA games ready to go there.

they sure found a way at Amazon to fuck it up in subtle and not-so-subtle ways

And here is my prejudice confirmed. Against better intuition, I actually downloaded this. Went into a random folder... dev\Code\Launcher\LinuxLauncher and opened a random file (Main.cpp) in an editor. Scrolled down a couple of lines. Right after the function IncreaseResourceMaxLimit you find what?

bool IncreaseSackSizeToMax()

Really, dude. Sack? Granted, you could as well call the function thingamagoop if that's what you wanted, and if people called it with that name it would work fine. But... you know, spellchecking?

What's most amazing is that the function is indeed being called from within RunGame, which means either the guy writing the function used auto-complete and blindly hit return after typing Inc without ever looking what was being inserted (as long as it gives no compiler error, all is good!), or he actually changed Stack into Sack because the compiler did show an error. I'm trying to imagine it was the first one. I'm really trying.

No seriously, I didn't spend hours looking for this... first random file I opened. So with that initial impression, I'm not going to look at another one.

I don't really thing a typo is a method name is a good measure of code quality . . . I'd rather inrease sack size than segfault >< The end user won't really know my sack massively increased but he would notice the crash!

I don't really thing a typo is a method name is a good measure of code quality

Well, no... but it is a measure for no peer review or any such thing happening (or, if it does happen, it's a sign that the process sucks). I'm not saying that I couldn't make a spelling error myself or haven't made one (I sure have, and not just once).

But remember, this is not your private toy code that you wrote in your bedroom at 3 in the morning. It's not some messenger chat with your buddy either.
This is (or should be, by all means) production quality code published by a multi-billion dollar company that is intended to be used by thousands of developers for software that millions of people will use. If you spot a rather obvious spelling error at the first glance, what bugs that are not immediately obvious can you expect?

Incidentially, that particular function isn't part of the official API that end-users, or rather developers, are to use (you could call it an "implementation detail" of the framework, if you wish). But imagine it was part of the API, it could very well be. There's no way you can change it after making it public (no good and straightforward way, anyway).

I'm willing to concede that maybe I found the single one spelling error in these 25k source files by accident, in the first file I opened. But that's about as likely as winning the superball. I should buy a ticket.

I don't really thing a typo is a method name is a good measure of code quality

Well, no... but it is a measure for no peer review or any such thing happening (or, if it does happen, it's a sign that the process sucks). I'm not saying that I couldn't make a spelling error myself or haven't made one (I sure have, and not just once).

But remember, this is not your private toy code that you wrote in your bedroom at 3 in the morning. It's not some messenger chat with your buddy either.
This is (or should be, by all means) production quality code published by a multi-billion dollar company that is intended to be used by thousands of developers for software that millions of people will use. If you spot a rather obvious spelling error at the first glance, what bugs that are not immediately obvious can you expect?

Incidentially, that particular function isn't part of the official API that end-users, or rather developers, are to use (you could call it an "implementation detail" of the framework, if you wish). But imagine it was part of the API, it could very well be. There's no way you can change it after making it public (no good and straightforward way, anyway).

I'm willing to concede that maybe I found the single one spelling error in these 25k source files by accident, in the first file I opened. But that's about as likely as winning the superball. I should buy a ticket.

It's more like annecdotal evidence and there are plenty of ways to fix this in future versions including but not limited to breaking changes, which aren't as much of a no go as everyone makes them to be when you see new engines adding new whole mechanisms that expect to change your whole workflow a patch note saying "a one letter change is required to fix a typo in all previous projects, you will be offered to have it automatically fixed in your whole project when you next launch the editor" seems like a really minor issue

Now if it ends being that the code is littered with such issues then yea that's a bad sign, but so far it's one even if you didn't look far enough to tell if it's just 1 or thouthands.

This is (or should be, by all means) production quality code published by a multi-billion dollar company that is intended to be used by thousands of developers for software that millions of people will use.

Yes, and there may also be a language barrier. Imagine a teleconference with some broken English on one end and crytek in Nottingham on the other end, the conversation going something like:

"users have reported a segfault in certain conditions. As our official outsourcing company for this part of the engine can you fix it"
"yes we fix, we make sefaul not happen by raise sack size OK?"

Just a thought...

If you're in any doubt to if this can and does happen read the last post on this page...

This is (or should be, by all means) production quality code published by a multi-billion dollar company that is intended to be used by thousands of developers for software that millions of people will use.

Yes, and there may also be a language barrier. Imagine a teleconference with some broken English on one end and crytek in Nottingham on the other end, the conversation going something like:

"users have reported a segfault in certain conditions. As our official outsourcing company for this part of the engine can you fix it"
"yes we fix, we make sefaul not happen by raise sack size OK?"

Just a thought...

If you're in any doubt to if this can and does happen read the last post on this page...

I don't think referencing a topic on coding horror is a good way to argue that this may not be a quality control issue.

Well, no... but it is a measure for no peer review or any such thing happening (or, if it does happen, it's a sign that the process sucks).

Cry is a culture based on big egos, power cliques, and intellectual dishonesty. Don't expect peer review to exist in such an environment.

It's also extremely multilingual, so spelling mistakes in English are to ve expected. And yes, they often propagate via autocomplete.

Except porting is probably close to free for all games that currently cryengine so that's a base of AAA games ready to go there.

Yep, all those many CryEngine games :lol:
Also, porting from one version of Cry to the next ie not simple, let alone a whole different fork from 2014.