Nvidia: OEMs to Blame for Re-branding GPUs

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

vinhcit

Distinguished
Apr 6, 2009
6
0
18,510
To 4745454b @ 03/05/2010 1:52 AM

Funny and likely scenario. But not the only possibility. It could have been:

OEM: Give us some new products to make our specs fresh.
NVIDIA: We don't have any yet. Working on Fermi. Just like the old ones.
OEM: K, how about give the old products new labels?
NVIDIA: OK!

I don't like rebranding but to be honest, it doesn't really significant affect anyone, enthusiasts (who know which products are which anyway) as well as normal users (who don't care, as to these users, midranged nvidia cards and midranged ati cards, regardless of generations and labeling, are practically indistinguishable).
 

1898

Distinguished
Oct 13, 2009
249
0
18,690
dang, thumbs up are limited to 20.

Topic:
Sole purpose of rebranding is to deceive the customers, nothing to argue about that. Deception is nothing new when it comes to Nvidia, BAA's disabled AA (for ATI cards only) is a good example of that.

Though, it doesn't seem to be illegal but as a customer I've got a choice and will surely show my appreciation for such actions.
 

It wasn't disabled because it was not there in the first place, anti-aliasing has always had to be added to games using the Unreal engine and ATi have always known that but in this instance they were too lazy to write their own code and wanted to use Nvidias to which the dev's legal dept said no.
 

idisarmu

Distinguished
Mar 23, 2008
511
0
18,980
I wouldn't mind rebranding the GPUs that much if they shrunk the die and overclocked it each time, but this is just ridiculous. 2900xt to 3870 was acceptable, 3870 to 4670 not so much (should have had ddr5 imo), but at least it was cheap and didn't require a power connector. Now if they were to make ANOTHER 320 SPU R600 based card, they BETTER make it at 40nm or less and have an 850 or greater core.
 

tayb

Distinguished
Jan 22, 2009
1,143
0
19,280
I'm not buying it. It's pretty easy just to say "No, we aren't going to rebrand this chip." I'd be willing to bet there are suppliers who would like ATI to rebrand their chips.
 

siuol11

Distinguished
Dec 16, 2009
77
0
18,630
[citation][nom]idisarmu[/nom]I wouldn't mind rebranding the GPUs that much if they shrunk the die and overclocked it each time, but this is just ridiculous. 2900xt to 3870 was acceptable, 3870 to 4670 not so much (should have had ddr5 imo), but at least it was cheap and didn't require a power connector. Now if they were to make ANOTHER 320 SPU R600 based card, they BETTER make it at 40nm or less and have an 850 or greater core.[/citation]

I think you are smoking a wee too bit much of the crack dude. You're probably making that connection because of the shader count I'm guessing? But The shader count is only a small part of a GPU, and each of these was made on a different process with different specifications (Driectx revisions give you some idea).
 

It's not the first time (and won't be the last) that a supplier has bent over to try and please their major customers, look back to 2006 with AMD and the AM2 Athlon CPU's that couldn't be got for love nor money because they were all going to Dell and HP leaving the retail channel dry for ages.
 
[citation][nom]ronch79[/nom]8-Series > 9 Series > GTX 200 Series > GTX 300 Series.Is this correct?[/citation]

The early 8 series used the G80 core. The late 8 series and the 9 series both used the G92 core, with the exception of the 9600GSO which I believe is a re-branded 8800gs G80 core. The G200 series used the GTX200 core, with the exception of the GTS250, which was a rebranded G92 core. The G300 series is re-branded and is either using a very crippled G200 core or the G92 core again I believe...not sure which.
 

The G300's and GT210, GT220 etc are using the 40nm die shrunk GPU, which is based on the G92 I believe.
 
Status
Not open for further replies.