If CoD doesn't use the full potential of today's hardware..

Originally Posted by David Hayward

I have no idea why they didn't test every configuration possible. Wait, yes I do. It isn't possible.

Well its possible now, since they sold us the game...

Salute !

[angrueo]

In this case, my opinion is that it is pretty clear that this game has been a forced release (no matter whether was finished or not). For today, game market stablish two or three profitable release dates over the year. The best date is in christmas, the second early october, and the third is mid-late march.

So, Ubi wanted this game released in the first quarter of fiscal year, on the basis that the main core of the game was done.

[Urufu_Shinjiro]

Originally Posted by recoilfx

The devs got hit with the epilepsy issue because the development was mismanaged. The team simply didn't have time to sort out everything that's been on Ubisoft's QA list when the chopping time came.

I don't mean that there are no talents in the team, but from a management point of view - they didn't make good decisions as to which features to pursue/refine in a given amount of time and funds.

They've bitten more than they can manage - time and money ran out. The blame lies, unfortunately, on Oleg and Luthier. They were the ultimate decision makers. I just hope that the team has enough cash flow from loyal fans to follow through.

I think you have it close but a little skewed. I think the game was pretty close to ready, in fact real close to where Oleg and crew wanted it to be for release, then Ubi sprang this epilepsy requirement on them and in any code this complex you're likely to break a whole bunch of stuff if you have to hack a feature in at the last minute. I think they were 100% forced to add this filter/make significant engine changes to meet the epilepsy requirement at the 11th hour and were given zero extra time to make it work. That would certainly explain Olegs youtube remarks about leaving game development, how frustrated would you be if you were nearing the release of all your labors and the result looked pretty damn good, then your contractual partner forced game breaking changes on you with no time to do it right or even make it work before release?

Originally Posted by angrueo

In this case, my opinion is that it is pretty clear that this game has been a forced release (no matter whether was finished or not). For today, game market stablish two or three profitable release dates over the year. The best date is in christmas, the second early october, and the third is mid-late march.

So, Ubi wanted this game released in the first quarter of fiscal year, on the basis that the main core of the game was done.

Yep, that's why Ubi forced Olegs hand regardless of the state of the product after the last minute epilepsy surprise.

[beazil]

Originally Posted by Urufu_Shinjiro

IYep, that's why Ubi forced Olegs hand regardless of the state of the product after the last minute epilepsy surprise.

IF true, that is sad news indeed. That still doesn't explain the lack of multi core or multi card rendering support by a piece of modern software that was engineered from the ground up to utilize modern computers for the purpose of BoB era flight simulation (otherwise why not just release another add on for IL2?).

It wasn't ready for release two weeks ago. It isn't now. I can see where it's going to be an absolutely glorious piece of work when it's finished. I don't even mind paying for the privalege of "beta testing" this monumental piece of work - but I do understand the frustrations of the developer, the customers and even the publisher - who has at some point to say just release it already.

This release has been something of a perfect storm for all involved. Which is a shame - because in the end I think we are going to get something I think we will all really be able to enjoy. S!

[recoilfx]

Originally Posted by Urufu_Shinjiro

I think you have it close but a little skewed. I think the game was pretty close to ready, in fact real close to where Oleg and crew wanted it to be for release, then Ubi sprang this epilepsy requirement on them and in any code this complex you're likely to break a whole bunch of stuff if you have to hack a feature in at the last minute. I think they were 100% forced to add this filter/make significant engine changes to meet the epilepsy requirement at the 11th hour and were given zero extra time to make it work. That would certainly explain Olegs youtube remarks about leaving game development, how frustrated would you be if you were nearing the release of all your labors and the result looked pretty damn good, then your contractual partner forced game breaking changes on you with no time to do it right or even make it work before release?


Yep, that's why Ubi forced Olegs hand regardless of the state of the product after the last minute epilepsy surprise.

I don't think Ubisoft is to blame for the Epilepsy issue. This has been a practice with Ubisoft for several years.

If Oleg signed on with a publisher, then it would be the Oleg's fault for not reading these requirements, and subsequently setting the wrong release date - again, a management mistake.

Luthier also had officially said that Ubisoft wasn't at fault.

Rushed out the door or not , they had years. Surely they would see that it runs like balls in this scenario and try to fix it. Nobodys system could really run it until last week with the NOSSA tweat . Now most people get great fps. A week after release and a customer found a fix.

[Heliocon]

I would like to note there is a few problems here:
1. Optimization, the current engine is currently rediculously inneficient not only in what it processes, but the fact that it is badly bottlenecked on even last gen hardware. So efficient use of more than one thread, SLI and Crossfire support etc. Anti-aliasing needs immediate loving too.
2. Once the above issues are addressed, they can start looking at how it will use the current gen high end hardware and software (that means 4 core CPUs efficiently and possibly 8 threads (althought the top end is 6 cores) and DX11-tesselation, direct compute etc) Also efficient use of 64bit.exe with memory over 4gb. Also needed is support for 3 way SLI and crossfire which is becoming very common (for example 3 460s).
3. Once the above is done they can start working on next gen hardware that will be released mid way-end to the end of this year, so by the time they get to it will probably be high end current gen (aka point 2) which is 8 core cpus, with 16 threads (AMD Bulldozer) and a possible 32nm 8 core and 16 thread intel cpu for 1366 socket. At the end of 2011 Nvidia is coming out with their brand new 28nm (current is 40nm I believe) GPUs, AMD is likely to do the same (AMD since ATI was bought out and they are phasing out the use of ATI brand name).
4. By this time then they can start looking at what is coming in the future

Originally Posted by Heliocon

I would like to note there is a few problems here:
1. Optimization, the current engine is currently rediculously inneficient not only in what it processes, but the fact that it is badly bottlenecked on even last gen hardware. So efficient use of more than one thread, SLI and Crossfire support etc. Anti-aliasing needs immediate loving too.
2. Once the above issues are addressed, they can start looking at how it will use the current gen high end hardware and software (that means 4 core CPUs efficiently and possibly 8 threads (althought the top end is 6 cores) and DX11-tesselation, direct compute etc) Also efficient use of 64bit.exe with memory over 4gb. Also needed is support for 3 way SLI and crossfire which is becoming very common (for example 3 460s).
3. Once the above is done they can start working on next gen hardware that will be released mid way-end to the end of this year, so by the time they get to it will probably be high end current gen (aka point 2) which is 8 core cpus, with 16 threads (AMD Bulldozer) and a possible 32nm 8 core and 16 thread intel cpu for 1366 socket. At the end of 2011 Nvidia is coming out with their brand new 28nm (current is 40nm I believe) GPUs, AMD is likely to do the same (AMD since ATI was bought out and they are phasing out the use of ATI brand name).
4. By this time then they can start looking at what is coming in the future

Pfiooooouuuw ! That means at least 400 gallons of Vodka ....

Salute !

[mazex]

He he, there is also something called "gold plating" regarding software development... That is when a developer puts down way to much time on some insignificant thing and makes it perfect beyond the demands of the customer while they miss doing a lot of other rather easy stuff that are important for the customer...

We have cockpits that are clickable which only a few percent of the users will use more than a few times (Olegs own words) - while we don't have force feedback which is a major issue for a good percentage of the core users, horrible input mappings etc... It sure will be fixed but it hurts seeing the control settings pages for a game of this calibre.

Still - as a software developer myself I can imagine the horror of tracking down bugs in a piece of software this complex where you don't know the hardware it will run on, or driver versions etc... As it is now it really is getting really enjoyable even on old mid range computers like mine (new one will be ordered when a bonus lands in two weeks).

Regarding bugs and bad performance... Just look at this video I did last night on my 4 year old computer (Core 2 Duo E8400@3.3 - GTX275 - 6GB memory and W7x64):

http://www.youtube.com/watch?v=qVw0pPopaQQ

That is with the latest patch and the nossao mod, I have no complaints about the performance there...

[Heliocon]

Originally Posted by mazex

He he, there is also something called "gold plating" regarding software development... That is when a developer puts down way to much time on some insignificant thing and makes it perfect beyond the demands of the customer while they miss doing a lot of other rather easy stuff that are important for the customer...

We have cockpits that are clickable which only a few percent of the users will use more than a few times (Olegs own words) - while we don't have force feedback which is a major issue for a good percentage of the core users, horrible input mappings etc... It sure will be fixed but it hurts seeing the control settings pages for a game of this calibre.

Still - as a software developer myself I can imagine the horror of tracking down bugs in a piece of software this complex where you don't know the hardware it will run on, or driver versions etc... As it is now it really is getting really enjoyable even on old mid range computers like mine (new one will be ordered when a bonus lands in two weeks).

Regarding bugs and bad performance... Just look at this video I did last night on my 4 year old computer (Core 2 Duo E8400@3.3 - GTX275 - 6GB memory and W7x64):

That is with the latest patch and the nossao mod, I have no complaints about the performance there...

The vid looks good - now what the irony here is that if its running well on that rig but someone with a far better rig (this applies to anything since whats high end soon becomes mid etc) does not see any improved performance, that is very worrying. How is performance around buildings, or with many planes in the air?
The reason the ssao off helps is because you are on a 275gtx, which is old dx10 hardware (its not dx10.1) which uses an updated shader and lighting pipeline very similar to DX11 (Shader model 4,4.5,5) while dx10 uses a pipeline closer to DX9 I believe (Shader model 2.0,2.5?,3.0).
The way it manages rescources is different, 10.1/11 brings a big performance boost if you run on more than 2 cores.
They are stuck in a tech rut...

Also your criticism of the nit picky perfectionism I think is 110% on the spot. You hit the nail on the head, they have focused alot of energy into perfectionism and very niche (within the already small niche) features instead of investing in the infastructure.