ATi same mistake as nVidia?
- Thread starter Chris123234
- Start date
Chalnoth said:
Ok, my bad.
Look at this.. OCS L2 Enhanced Radeon 9700Pro : http://www.monster-hardware.com/reviews/ocsystemati9700 1.htm
The same still applies tho.. Meaning you probably can't cool an Ultra FX with a zalman, but with the R9700 (and overclocked) you can.
If when the R9900 is released, and some vendors have solutions with a zalman.. that will certainly raise a few eyebrows
demonic said:But, what I mean't. Is if ATi *released* the Rv350, with just a heatsink. Just imagine the picture it would send to everyone?
In the blue corner we have ATi, with just a heatsink.
In the red corner we have Nvidia, with the dustbuster!!
Now you get what I mean
It would surprise me to find that there were still people around who didn't already have that picture firmly engraved in their gray matter...
I know I certainly do. With a good design you should be able to take it down a size and have it run on less voltage and cooler than it did on the former process--which means you can then ramp it up in MHz speed and increase the voltage a tad and still come out with an overall thermal profile close to what you had on the larger process at slower MHz speeds and less voltage. IMO. You can foul this up if you abuse the shrink process and tack on too much additional circuitry, though--which is maybe what nVidia did.WaltC said:
Or you can get burned by a process that isn't ready for prime time. While I can't say for sure that NVIDIA's engineers didn't bungle something, its been "the word on the street" that TSMC's .13u process is still not doing so well.
WaltC said:
No, the CEO said that.
Or maybe he was saying stuff to pump up his upcoming product (which is in .13) and denigrating the competitors product by saying nothing as wizbang could be made in .15.
Joe DeFuria
Legend
No matter how you slice it:
The NV30 "launch" is a f'n MESS.
And nVidia, AS A COMPANY, looks bad because of it. Whether Huang doesn't have a clue, or he's jut pretending not to have one for the sake of "hype", it reflects poorly on nVidia.
The NV30 "launch" is a f'n MESS.
And nVidia, AS A COMPANY, looks bad because of it. Whether Huang doesn't have a clue, or he's jut pretending not to have one for the sake of "hype", it reflects poorly on nVidia.
Joe DeFuria said:
Oh, yea--I can't remember anything this bad. nVidia's well on its way to usurping 3dfx's record for "the worst-botched 3D card/chip release in history".....! The company has done nothing but foul this up for months--quite unbelievable. If I didn't think otherwise I might be tempted to think the remnant of 3dfx within nVidia had Trojan-horse'd nVidia, but good....
Likely though that it's something much less spectacular--like them tripping over their own hubris. You might also think the R300 rattled them senseless. Who knows.As I said back when the nv30 hype was building, I have hope that the lessons nVidia takes away from this will change their PR and marketing strategies.
That is assuming they conclude there was problem in the nature of their marketing and PR conduct, instead of believing there just wasn't enough of it.
Let's see how this year shapes up, and I hope their engineering is up to competing with ATI so that the alternative I'm hoping for is more appealing to them...
That is assuming they conclude there was problem in the nature of their marketing and PR conduct, instead of believing there just wasn't enough of it.
Let's see how this year shapes up, and I hope their engineering is up to competing with ATI so that the alternative I'm hoping for is more appealing to them...
Chris123234
Regular
Maybe I was right
Upcoming ATI-Desktop-Graphicscards
Radeon 9800 0,15 µm 375 MHz DDR-I 128 MByte 350 MHz
Radeon 9600 Pro 0,13 µm 400 MHz DDR-II 128 MByte 400 MHz
Radeon 9600 0,13 µm 300 MHz DDR-II 128 MByte 300 MHz
Radeon 9200 0,15 µm 275 MHz DDR-I 64 MByte 275 MHz
the 9600 (RV350) has a higher clock rate and faster memory
http://www.guru3d.com/comments.php?category=1&id=1326
Upcoming ATI-Desktop-Graphicscards
Radeon 9800 0,15 µm 375 MHz DDR-I 128 MByte 350 MHz
Radeon 9600 Pro 0,13 µm 400 MHz DDR-II 128 MByte 400 MHz
Radeon 9600 0,13 µm 300 MHz DDR-II 128 MByte 300 MHz
Radeon 9200 0,15 µm 275 MHz DDR-I 64 MByte 275 MHz
the 9600 (RV350) has a higher clock rate and faster memory
http://www.guru3d.com/comments.php?category=1&id=1326
Nebuchadnezzar
Legend
You won't understand it, won't you?
Chris123234
Regular
"0,13 µm 400 MHz DDR-II 128 MByte 400 MHz " looks like a normal FX 5800 to me
JF_Aidan_Pryde
Regular
WaltC:
That Trojan theory was my idea!! MINE!!
Seriously though, it's simple. NVIDIA has been missing product cyles ever since the NV15. This time a competitor made them pay.
That Trojan theory was my idea!! MINE!!
Seriously though, it's simple. NVIDIA has been missing product cyles ever since the NV15. This time a competitor made them pay.
JF_Aidan_Pryde said:WaltC:
That Trojan theory was my idea!! MINE!!
*chuckle*... 8)
I've always thought nVidia's idea here was both logical and stupid. Logical because it helped them cook 3dfx's goose (kind of an "arm's race" manifesto). Stupid because it was obviously unsustainable from the start. You are right, they've missed several of their self-ordained cycles, most of them I think after 3dfx dropped out of the picture. But I really think what ATI did here with R300 goes far beyond nVidia's product cycle schedule--I think ATI just leapfrogged 'em, plain and simple--kind of like AMD did Intel back in '99 (although I'd end the comparison there because ATI and nVidia are much closer in terms of resources than are AMD and Intel.) nVidia played a good come-from-behind game with 3dfx, but ATI is no 3dfx, and as such I think the kind of lead ATI has will be much more difficult for nVidia to overcome than the situation was with 3dfx. I think 2003 will cement in the trends with respect to these issues that we are likely to see for at least serveral more years after.
Similar threads
