Nvidia GT200b rumours and speculation thread

inq just said the 200b is in for another spin

270/290 in deep trouble

IT LOOKS LIKE Nvidia's run of bad luck is continuing. The latest word is that the GT200b is in for another respin, that would be the third. This means the 270 and 290 are effectively delayed until February, and the GX2 part is quite dead.

They're further referring to another "Ultra(rare)" release this month for the Christmas Halo.
 
inq just said the 200b is in for another spin

270/290 in deep trouble

seems plausible, since nehalem can use sli with intel chipsets it would have made sense for new chips to launch just around it. Absolutely no news on coming derivatives/shrinks from green camp does seem to imply to me that the respins/shrinks are not quite working out for them.
 
I really hope for Nvidia the Inq is wrong on that (something like their 8800GTX sucks or Nvidia leaves chipset market news).
 
I`m afraid that it could be true. If not then why there is still no 55nm refresh from NV? They could have some serious problems with it. It was prepared for September/October time, then November and now next year.

GT200B looks like r600 at now:( (still delayed)
 
GT200 already pretty much matched R600 in terms of lateness.
The difference is that ATI didn't have anything to take advantage of GT200's lateness, and NVidia didn't really lose much either. G92 was selling with healthy margins all the way up to GT200's launch, and the 9800GTX was scheduled to drop down to $299 when GT200 arrived. With an earlier launch, it's hard to say whether the GT200 cards would give better margins than a 9800 GX2 or $400 9800GTX.

RV770 is what messed everything up for NVidia, not GT200's tardiness.
 
The difference is that ATI didn't have anything to take advantage of GT200's lateness, and NVidia didn't really lose much either. G92 was selling with healthy margins all the way up to GT200's launch, and the 9800GTX was scheduled to drop down to $299 when GT200 arrived. With an earlier launch, it's hard to say whether the GT200 cards would give better margins than a 9800 GX2 or $400 9800GTX.

RV770 is what messed everything up for NVidia, not GT200's tardiness.
We've been here already:

http://forum.beyond3d.com/showthread.php?p=1206533#post1206533

My opinion hasn't changed :p

8800GT was <$350 and 8800GTS-512 was <$400 at Christmas 2007:

http://www.anandtech.com/video/showdoc.aspx?i=3175

8800GTX/U prices had started their crash.

If GT200 had released on time then I think RV770's launch in June would have elicited a "so-what?" shrug - "at best equal performance" at a slightly cheaper price. GT200's price would have had time to ease so RV770's price shock wouldn't have been anything like as pronounced.

GT200 being on time would have also made NVidia's shenanigans with G92's history of band naming and re-re-naming less prominent - enthusiast forum sentiment consisted of "stop taking the piss and release something we want to buy".

Jawed
 
Not if they were supposed to be intro'ed in the same timeframe. GT200 (or G90?) as high end substitute of G80 while G92 would transition G80's performance into the mainstream segment.

Well, Roadmaps. :)
 
Last edited by a moderator:
Ah this is a bummer but then again I can go ahead and build a top of the line Nehalem system with ATI cards, relegate my current system to be hooked up to my soon to be 60 inch plasma as a media server type box and carry on...I really wanted new vid cards! :(
 
Ah this is a bummer but then again I can go ahead and build a top of the line Nehalem system with ATI cards, relegate my current system to be hooked up to my soon to be 60 inch plasma as a media server type box and carry on...I really wanted new vid cards! :(

Im sure a pair of AMD 4870X2s would ease the pain somewhat! :)
 
I may get shouted down for this but i'd much prefer a pair of 280GTX's than a pair of 4870X2's.

Quad fire is only going to bring compatability problems, input lag, and a minimal performance boost over a single X2 in many cases.

280SLI should give much more consistent and reliable performance with half the input lag. And its cheaper too!
 
Im sure a pair of AMD 4870X2s would ease the pain somewhat! :)

There is a slight catch to that unfortunately Squilliam and that is that the motherboard I have only does Nvidia video cards in SLI. I have the option of building a Nehalem based machine with an Intel mobo thus becoming agnostic to whatever video cards, but thats just extra money I dont want to spend right now especially considering how fast my rig is.

I may get shouted down for this but i'd much prefer a pair of 280GTX's than a pair of 4870X2's.

Quad fire is only going to bring compatability problems, input lag, and a minimal performance boost over a single X2 in many cases.

280SLI should give much more consistent and reliable performance with half the input lag. And its cheaper too!

And yes I have been hearing about some issues with the quad crossfire. It still seems to me that the drivers are not up to speed. I dont know but it sounds like the same stuff that the GX2 Nvidia quad sli folk experienced. Just not worth it IMHO>
 
From a newer XFX XXX:


... so it seems there is a better version of GT200A out, since usual GTX 280 an older XXX used G200-300-A2?
 
55nm GT200 in Quadro FX 5800?

Latest Quadro FX graphics card has 4GB of RAM

Quadro FX 5800
  • 52 billion texels/s
  • 4GB GDDR3 @ 800MHz
  • 189W TDP
GTX280
  • 48.2 billion texels/s
  • 1GB GDDR3 @ 1107MHz
  • 236W TDP

So there's a slightly increased core clock - 650MHz if my maths are correct - and a much lower TDP. Sure the GDDR3 is clocked lower but there's four times(!) as much of it.

I posted this in the GT206/212 thread also since I don't know what's what with NVIDIA codenames anymore :p.
 
Yeah I think we'd concluded a while back that the 55nm refresh was already in production for non-consumer cards - so there's a very strong likelihood this is indeed 55nm.

Jawed
 
If the quadro shrink is out, then I wonder why it hasn't shown up in consumer cards yet. I wonder if the inquirer's suggestion of slipstreaming 55nm parts into existing models is true.
 
Back
Top