GeForce 9900 GTX & GTS Slated For July Launch

Printable View

Show 100 post(s) from this thread on one page

05-26-2008, 08:47 PM
LordEC911

Quote:

Originally Posted by zerazax

And it looks like the ratio is 24 SP's per 8 TMU's as well which isn't quite the 2 to 1 for G92 but better than the 4 to 1 on G80 (though it was 32TA, 64TF)

I'm pretty sure that it is 40TMUs and 80TFUs, ala G80 style.
05-26-2008, 09:11 PM
Loque

will we get some benchmarks this week ?
05-27-2008, 12:28 AM
YukonTrooper

Quote:

Originally Posted by Loque

will we get some benchmarks this week ?

I'm going to Jamaica tomorrow for two weeks with my brothers, so I know I'm going to come back to all kinds of goodies, maybe even a launch!

BTW, I just finished playing Gears of War on hardcore. Excellent game. Noticed your avatar.
05-27-2008, 01:47 AM
Nuker_

Quote:

Originally Posted by Yukon Trooper

I'm going to Jamaica tomorrow for two weeks with my brothers, so I know I'm going to come back to all kinds of goodies, maybe even a launch!

BTW, I just finished playing Gears of War on hardcore. Excellent game. Noticed your avatar.

Isn't his avatar UT3 though?
05-27-2008, 03:35 AM
Nedjo

i wonder how badly will this reflect on GT200 since it's gigantic size of 576mm^2:

http://www.reuters.com/article/techn...080527?sp=true
05-27-2008, 03:38 AM
Shintai

Quote:

Originally Posted by Nedjo

i wonder how badly will this reflect on GT200 since it's gigantic size of 576mm^2:

http://www.reuters.com/article/techn...080527?sp=true

The same as if it was a 100mm2 chip. TSMC seems to want to raise the price primary on 45nm. And else overall on the GP process. Size doesnt matter, what process you use does. TSMC just saw how much it starts to hurt being in the big boys league on process advancement.
05-27-2008, 05:07 AM
NH|Delph1

Quote:

Originally Posted by natty

whats the story with the 260 shader clock, some sources say 1240 others say 999

It spawned from an error in an article over at TheInq. They wrote 999MHz 896MHz GDDR3 memory, when they meant to write 999MHz 896MB GDDR3 memory.

The shaders are clocked at ~1240MHz.

//Andreas
05-27-2008, 06:59 AM
Nedjo

Quote:

Originally Posted by Shintai

The same as if it was a 100mm2 chip. TSMC seems to want to raise the price primary on 45nm. And else overall on the GP process. Size doesnt matter, what process you use does. TSMC just saw how much it starts to hurt being in the big boys league on process advancement.

read again, don't made up things... BTW TSMC doesn't have 45nm process, they have 40nm...
05-27-2008, 07:16 AM
NH|Delph1

Quote:

Originally Posted by Nedjo

read again, don't made up things... BTW TSMC doesn't have 45nm process, they have 40nm...

40nm is a half-node, how can they do a 40nm half-node without the 45nm node?

//Andreas
05-27-2008, 07:27 AM
Shintai

Quote:

Originally Posted by Nedjo

read again, don't made up things... BTW TSMC doesn't have 45nm process, they have 40nm...

You might want to tell TSMC that, plus what Andreas said
http://www.tsmc.com/english/b_techno...10101_45nm.htm

And yes I read it, or are you claiming AMD and nVidia wouldnt be in the same boat? Perhaps you aint quite sure what you posted :p:
05-27-2008, 07:53 AM
Nedjo

Quote:

Originally Posted by Shintai

You might want to tell TSMC that, plus what Andreas said
http://www.tsmc.com/english/b_techno...10101_45nm.htm

And yes I read it, or are you claiming AMD and nVidia wouldnt be in the same boat? Perhaps you aint quite sure what you posted :p:

usually you are putting much more effort in Googling when you're trying to back up your statements! Why haven't you done same this time? Wait I know: 'cos you're wrong!

1. TSMC will skip 45nm and go diretly to 40nm: http://www.semiconductor.net/blog/27.../10025801.html

2. I know it's hurting you, but you must face reality: AMD and NVIDIA are not in the same boat with upcomming generation of GPU's! Why? Simple: let's say that wafer price for both of them is equal and AMD with their RV770 is getting twice as much chips comparing to GT200 in ideal yield situation! What's realistic to expect is that AMD will have better yields due to smaller chip... so this inflation of wafer price will have more severe consequences on NVIDIA in comparison to AMD.

BTW congrats on your 4K post ;)
05-27-2008, 07:55 AM
Luka_Aveiro

Quote:

Originally Posted by Shintai

You might want to tell TSMC that, plus what Andreas said
http://www.tsmc.com/english/b_techno...10101_45nm.htm

And yes I read it, or are you claiming AMD and nVidia wouldnt be in the same boat? Perhaps you aint quite sure what you posted :p:

You are wrong :p:
05-27-2008, 08:14 AM
jas420221

From your link....

Quote:

TSMC and the Reverse Temperature Effect
April 30, 2008
Taiwan Semiconductor Manufacturing Co. Ltd. (TSMC) held its annual technology symposium in Austin Tuesday (April 29), with much of the attention on the foundry’s 40 nm technology. TSMC would prefer that its leading edge customers go directly from 65 to 40 nm design rules, making 40 nm much more than an afterthought 0.9× linear shrink. In fact, TSMC will skip 45 nm and only offer 40 nm for the general (G) and (LPG) processes, with 45 nm and 40 nm offerings for the low power (LP) process which Qualcomm Inc. and others use.

Seems like they are using 40nm but not for PC processors?
05-27-2008, 08:19 AM
Shintai

Quote:

Originally Posted by Nedjo

usually you are putting much more effort in Googling when you're trying to back up your statements! Why haven't you done same this time? Wait I know: 'cos you're wrong!

1. TSMC will skip 45nm and go diretly to 40nm: http://www.semiconductor.net/blog/27.../10025801.html

2. I know it's hurting you, but you must face reality: AMD and NVIDIA are not in the same boat with upcomming generation of GPU's! Why? Simple: let's say that wafer price for both of them is equal and AMD with their RV770 is getting twice as much chips comparing to GT200 in ideal yield situation! What's realistic to expect is that AMD will have better yields due to smaller chip... so this inflation of wafer price will have more severe consequences on NVIDIA in comparison to AMD.

BTW congrats on your 4K post ;)

If you need twice as many ATI chips to be at the performance of the nVidia chips. Then you still are in the same boat.

For 45nm, they still developed and use 45nm for LP. But as they wrote, 45nm dont hold enough benefits so they go for 40nm. Yet more waitingtime.
If anything its just a show of things aint going as they should.
05-27-2008, 09:49 AM
YukonTrooper

Quote:

Originally Posted by Nuker_

Isn't his avatar UT3 though?

Maybe you're right. Just looks a lot like Dom from GoW. I've never played UT3.
05-27-2008, 10:17 AM
Nedjo

Quote:

Originally Posted by Shintai

If you need twice as many ATI chips to be at the performance of the nVidia chips. Then you still are in the same boat.

Real world performance of these chips hasn’t got anything to do with the subject of how much do you harvest them from the wafer…

Quote:

For 45nm, they still developed and use 45nm for LP. But as they wrote, 45nm dont hold enough benefits so they go for 40nm. Yet more waitingtime.
If anything its just a show of things aint going as they should.

Only thing that’s publicly available about TSMC’s nex-gen manufacturing process is that they are very satisfied with the progress and are even so confident in the quality that they dare to compare it with Intel’s High-K process.
http://www.beyond3d.com/content/news/636
05-27-2008, 11:05 AM
Shintai

Quote:

Originally Posted by Nedjo

Real world performance of these chips hasn’t got anything to do with the subject of how much do you harvest them from the wafer…

Thats like comparing a G94 etc chip with a RV770. Performance matters when you need 2 AMD chips to compete with 1 nVidia chip. Thats how it works. Else you are comparing the wrong chips.

And again, even if TSMC incrased the price for nVidia only with 20% it would still leave AMD/ATI in the red and nVidia shovelling in the cash as usual. Your point is useless. Any price increase hits equally on the 2. If nVidias chips get 5% more expensive, so will AMDs.
05-27-2008, 01:25 PM
t800

When these new 9900 will be available on stores?...july?
05-27-2008, 01:34 PM
BenchZowner

Quote:

Originally Posted by t800

When these new 9900 will be available on stores?...july?

June 17th.

It's not 9900 though.
05-27-2008, 01:53 PM
AliG

how much will the 9900gtx cost? I know for sure I can't afford the gtx 280, so its either that or a 4870 for me
05-27-2008, 01:58 PM
Shintai

Quote:

Originally Posted by AliG

how much will the 9900gtx cost? I know for sure I can't afford the gtx 280, so its either that or a 4870 for me

260GTX you mean?
05-27-2008, 01:59 PM
BenchZowner

Can't really tell AliG
Depends on many many variable factors ( variable as factors can change ).
Normally it shall be launched at 349$.
Don't take this statement to the bank though :p:
05-27-2008, 02:05 PM
perkam

HD 4870 512MB GDDR5 $349
9900GTX (G92b) 512MB GDDR3 $299
HD 4850 512MB GDDR3 $229
9800GT 512MB (G92) $199

Somewhere along there. Though, the HD 48xx should be better than both of Nvidia's counterparts.

Perkam
05-27-2008, 02:16 PM
GoldenTiger

I want to know the GTX 260/280 pricing for sure haha... hoping I can afford the 280 :D.
05-27-2008, 02:20 PM
BenchZowner

Quote:

Originally Posted by GoldenTiger

I want to know the GTX 260/280 pricing for sure haha... hoping I can afford the 280 :D.

Already mentioned.
The 280 shall be priced at... only 649$

Show 100 post(s) from this thread on one page