NVIDIA shows signs ... [2008 - 2017]

Status
Not open for further replies.
Dave Baumann, from your point of view, is ATI on the same level with nvidia concerning dev support?

In the past there were always messages, that the dev support of ATI is not as good as it should be, especially compared to nvidia.

Messages from who?

I played MMOG and I know ati been into those as the cards from the red team always have been to my knowledge to prefer when playing MMOG.
There are a few MMOG that came out where Nvidias cards simply sucked really with stuttering. eq2 was one.
Vanguard for example was used with ati cards in development, to bad the game sucked in how it was developed, Microsofts Vista was almost as bad.

so, messages from who?
Nvidia?
 
AA, PhysX, DX10, etc. Is that really hard to understand?

I think Dave was responding there to how impressively one-sided your arguments are (a la NV makes your games better, AMD just engages in marketing).

And the only DX10.1 game which has some kind of a noticable DX10.1 implementation got it via a patch after everyone on Earth finished playing it (Clear Sky).

HAWX anyone? Not a major release, but certainly a game that showed very substantial AA performance gains from its 10.1 support. AMD was making that game better, NV was just marketing against 10.1 support. Oh my.

API has nothing to do with this. At all. It's all about the hardware and selling this hardware. Nor AMD neither NV sell APIs.

So both graphics firms just pull their targetted feature sets for each generation out of thin air then? And what's the point of having a commonly supported API used as a technology inflection point for PC hardware if devs are promoted to fragment their support between competing IHVs? Does PC gaming in general benefit from that? Probably not, and if so then such advice from any devrel team strikes me as short-sighted when your parent company is in the business of selling discrete add-in boards used mainly for gaming. Like someone wrote upstream, we really should be condemning anything that fragments or damages the PC as an open gaming platform (and, yes, I realize that the wintel platform isn't truly open but since the vast majority of PC game devs use it there's no point in havering over semantics).
 
but also spent time on thier own general DX10 path, meaning that everyone gets benefit

Well done dave
/me salutes


And the only DX10.1 game which has some kind of a noticable DX10.1 implementation got it via a patch after everyone on Earth finished playing it (Clear Sky).

assassins creed also got a patch, there were a lot of rumours it got pulled because of pressure from nv
 
Last edited by a moderator:
I think Dave was responding there to how impressively one-sided your arguments are (a la NV makes your games better, AMD just engages in marketing).

AMD is very selective on what games they help out on ;), thier Dev Rel is more geared towards immediate and tangible benifits ASAP, nV's its more of fostering type and they help out quite a few developers from large to small, of course they too look at $ but they also know small fish sometimes can make a hit. And support wise, I don't know anyone that can say ATi, even AMD on the CPU side, has good support comparitive to Intel or nV.

HAWX anyone? Not a major release, but certainly a game that showed very substantial AA performance gains from its 10.1 support. AMD was making that game better, NV was just marketing against 10.1 support. Oh my.


Performance with later nV drivers the performance increase of the dx10.1 path didn't make a win for ATI cards.

http://www.computerbase.de/artikel/hardware/grafikkarten/2009/test_5_spiele_direct3d_101_test/5/
 
Last edited by a moderator:
This thread smells worse than the ace socks I got for Christmas last year, that I love to wear for days on end.

An upturn in the signal:noise would be appreciated.
 
Mirror's Edge is a UE3 title that uses exclusively DX9 and has AA support.


That's more due to its illumination solution (prebaked + shadowing) though, exhibiting far less artifacts than a normal UE3 game would show.
 
8800GTX/Ultra Failures.

I'm not sure what's going on here but the past few weeks have seen an alarming rate of 8800GTX/Ultra cards passing to greener pastures.

I'm not convinced this is related to bumpgate (yet) but it seems that G80 is hitting "old age".

(sites in dutch)
http://gathering.tweakers.net/forum/list_messages/1367271
http://gathering.tweakers.net/forum/list_messages/1366384
http://gathering.tweakers.net/forum/list_messages/1357213
http://gathering.tweakers.net/forum/list_messages/1358285
http://gathering.tweakers.net/forum/list_messages/1333851
http://gathering.tweakers.net/forum/list_messages/1329787
http://gathering.tweakers.net/forum/list_messages/1357023
http://gathering.tweakers.net/forum/list_messages/1351450

3870464543_b5a27d36ec_o.jpg


Most of them seem to exhibit normal failure behavior (artifacts, no video.)

Anyone having any experiences / numbers besides normal technical descriptions like "a lot" etc?
 
One of my FX 5900s died last week. Does that count...? ;)

Methinks this thread should be retitled "B3D members show signs of strain".
 
Last edited by a moderator:
The only cards that have ever died on me were ATI cards does that mean something?



Yeah it means I bought more ATI cards than Nvidia cards. That and I was unlucky I guess since their warranties are/were crap.
 
Feel free to put whatever you want in an easy bake. Me I will leave my g80 in my computer thanks...And my laptop with its nvidia GPU as well that is still fine after almost two years. *knocks on wood*.
 
One of my FX 5900s died last week. Does that count...? ;)

Methinks this thread should be retitled "B3D members show signs of strain".


I am beginning to think the same thing. Are we actively trying to find things wrong now? lol
 
Yeah, no such things on English forums (that was yesterday.)

Serious guys, look around or ask somewhere and this summer seems to be stressing them too much. Conveniently just passed european warranty they're baking targets now.

For every anecdotal find of an 8800 failure I can find one for an AMD part

I'm sure we would be able to find evidence of R600 failures too if anybody still actually owned those things ;) In any case, I'm not sure how scraping the internet for reports of failed hardware is evidence of a pattern. I can do the same for motherboards, memory sticks, hard drives etc.

Why are you so desperately searching for this anyway? If there was a pattern of failures it would emerge naturally.
 
Status
Not open for further replies.
Back
Top