I would like to say thank you to everyone at Toms for the years of great info and fun and help that they have provided. You Guys rule. It's too bad that so many people have a problem. But it's their problem and certainly not yours. To all my fellow Toms readers. Just enjoy what your givin and quit looking at the glass as half empty. If you don't want to see anything about Fermi then just don't click the link to read it. There is truly no point in complaining in Toms forum. It's a slap in the face to the individuals that work hard to bring it to the people who WANT to see it. Cheers.
- Status
Looks badass. I honestly pray it performs amazingly to go buy me one already. Come on Nvidia, dont freaking let go cause its also our only hope to balance high end GPU prices again or ATI will be shooting higher soon. ( Plus the fact ill be forced to pay for it since ill have to purchase a 5970 😛 )
brisingamen
Distinguished
- Feb 3, 2009
- 201
- 0
- 18,680
fermi is a flop, the real card that should be hyped up is the asus radeon 5870 rog 2gb version with a custom pcb like the lightning by msi. . .
the 58xx series of chips is an all around superior design.
the 58xx series of chips is an all around superior design.
englandr753
Distinguished
- Sep 4, 2007
- 2,653
- 8
- 21,115
Well, Nvidia is totally capable of coming back and they likely will even if its after Fermi. I still don't regret paying full price for the 8800GTX ACS3. You can't complain when you buy a card and its king of the hill for 3 years running. I love both ATI and Nvidia and I want them to continue to take the performance crown from each other. Its fun to watch and the customers always win no matter whos on top for the moment. It is good to see ATI/AMD in the middle of the winners circle for a change and I'm not sure I want them to lose the performance crown so soon. They have had a long hard struggle with the merger and getting wholloped by Intel. Go ATI.... er um Nvidia....lol, go both ATI and Nvidia!!!! =P
- Jul 1, 2008
- 4,491
- 0
- 22,780
[citation][nom]kyosho[/nom]I would like to say thank you to everyone at Toms for the years of great info and fun and help that they have provided. You Guys rule. It's too bad that so many people have a problem. But it's their problem and certainly not yours. To all my fellow Toms readers. Just enjoy what your givin and quit looking at the glass as half empty. If you don't want to see anything about Fermi then just don't click the link to read it. There is truly no point in complaining in Toms forum. It's a slap in the face to the individuals that work hard to bring it to the people who WANT to see it. Cheers.[/citation]
You are most welcome.
And thank you for the kind words and appreciation; they make it all worth the work.
You are most welcome.
And thank you for the kind words and appreciation; they make it all worth the work.
steiner666
Distinguished
- Jul 30, 2008
- 369
- 0
- 18,780
wow, it bears a striking resemblance to every other nvidia card in the past few years, whodathought?
It's ok to make ppl wait a long time for a new prodct OR to release a product thats barely ahead of the competition. But to do BOTH is just... UGH! I can't believe I've been putting off getting a 295 for about a year because of the hopes i had for these... :sad face:
It's ok to make ppl wait a long time for a new prodct OR to release a product thats barely ahead of the competition. But to do BOTH is just... UGH! I can't believe I've been putting off getting a 295 for about a year because of the hopes i had for these... :sad face:
Hmmm heat sink pipes hanging out the side mees thinkin they had no where else to puts them, don't dangle any wires on them, I think this is going to add to climate change for real this time.
dbaker9549
Distinguished
- Dec 1, 2009
- 6
- 0
- 18,510
what are the specs..no need to post anything till spec are out....so what good is a lambo with no engine...
nottheking
Distinguished
- Jan 5, 2006
- 1,456
- 0
- 19,310
It's a shame... I'll give nVidia credit for making what looks like one of the most wicked cooler designs I've EVER seen on a video card; possibly the best. (I've been a bit partial to the 7800GTX 512, I'll admit, as impractical as it was)
However, it looks like, as predicted, it's gonna be a bit of a bust; the 480 looks like paying almost the price of a 5970, with its TDP, for... Barely more than a 5870. Given that 40nm JUST became available for those outside of Intel, we're looking at the 480 being "that's all, folks" from the green team until probably early 2011. (possibly as early as this November)
All told, it seems an incredibly bad miss. Which actually is pretty bad, since it looks like AMD won't have to cut prices at all; even at their already above-MSRP states on NewEgg, they're still besting Nvidia on price/performance by a good margin, AND they're making a killing, while Nvidia undoubtedly will barely have a profit margin with these numbers.
So everyone say hello to half a year of market stagnation!
[citation][nom]liquidsnake718[/nom]I wonder why it was never followed up for the 9xxx series.......[/citation]
That's because in order to have a wider interface, you need more pins. And to connect more pins to a chip... You need more space along the chip to attach them to. In other words, there's a minimum GPU die size for each memory width; about 200mm² is needed for a 256-bit memory interface, and EVERY GPU with a 384-bit interface or higher has been at least 400mm². Hence, the original G80 was a massive 484mm², and could easily fit a 384-bit interface. but with the G92's die shrink, the size cut down to only 324mm², too small to actually fit anything larger than a 256-bit interface.
This is the same reason RV670, RV770, and RV870 all have 256-bit interfaces; if even tiny chips could have wide ones, heck, every card would be 512-bit, limited only by how many RAM chips they could afford/cram onto the board. But in truth, memory interface width is only one factor in memory bandwidth; one can compensate for a narrower interface with higher-clock memory, and likewise, compensate for low-clock memory with a wider interface. Hence why nVidia never used GDDR4 and was late to the GDDR5 party, as they had wider interfaces.
However, it looks like, as predicted, it's gonna be a bit of a bust; the 480 looks like paying almost the price of a 5970, with its TDP, for... Barely more than a 5870. Given that 40nm JUST became available for those outside of Intel, we're looking at the 480 being "that's all, folks" from the green team until probably early 2011. (possibly as early as this November)
All told, it seems an incredibly bad miss. Which actually is pretty bad, since it looks like AMD won't have to cut prices at all; even at their already above-MSRP states on NewEgg, they're still besting Nvidia on price/performance by a good margin, AND they're making a killing, while Nvidia undoubtedly will barely have a profit margin with these numbers.
So everyone say hello to half a year of market stagnation!
[citation][nom]liquidsnake718[/nom]I wonder why it was never followed up for the 9xxx series.......[/citation]
That's because in order to have a wider interface, you need more pins. And to connect more pins to a chip... You need more space along the chip to attach them to. In other words, there's a minimum GPU die size for each memory width; about 200mm² is needed for a 256-bit memory interface, and EVERY GPU with a 384-bit interface or higher has been at least 400mm². Hence, the original G80 was a massive 484mm², and could easily fit a 384-bit interface. but with the G92's die shrink, the size cut down to only 324mm², too small to actually fit anything larger than a 256-bit interface.
This is the same reason RV670, RV770, and RV870 all have 256-bit interfaces; if even tiny chips could have wide ones, heck, every card would be 512-bit, limited only by how many RAM chips they could afford/cram onto the board. But in truth, memory interface width is only one factor in memory bandwidth; one can compensate for a narrower interface with higher-clock memory, and likewise, compensate for low-clock memory with a wider interface. Hence why nVidia never used GDDR4 and was late to the GDDR5 party, as they had wider interfaces.
dragonsqrrl
Distinguished
- Nov 19, 2009
- 1,280
- 0
- 19,290
[citation][nom]nottheking[/nom]However, it looks like, as predicted, it's gonna be a bit of a bust; the 480 looks like paying almost the price of a 5970, with its TDP, for... Barely more than a 5870.[/citation]
The HD5970 is currently going for around $700-$750, and that's if you can actually find one available. As far as I've seen, the 5970 has been very difficult to find since its release. Leaked prices have shown the GTX480 to be in the $500 price range, so I don't think I would consider the GTX480 "almost the price of the 5970", a $700 card. I would be more inclined to call it two entirely different price points.
And TDP is not determined by the wattage that can be 'potentially' supplied to the GPU. It's the wattage the GPU consumes at max load, its thermal envelope. I'm not sure why so many people have been saying this (because the GTX480 uses six + eight pin PCI-E power connectors it therefore uses the maximum amount of wattage that this configuration can supply, or it therefore uses the same amount of wattage as the HD5970) but this is not necessarily true. I don't know maybe the intent is to mislead those who don't know any better?
The GTX280 used the same connector configuration, and it consumed around 225 watts. The addition of the eight pin was basically meant as a safety against under powering the card. Assuming that the GTX480 consumes 295 watts is an over simplification, and well... misleading.
The HD5970 is currently going for around $700-$750, and that's if you can actually find one available. As far as I've seen, the 5970 has been very difficult to find since its release. Leaked prices have shown the GTX480 to be in the $500 price range, so I don't think I would consider the GTX480 "almost the price of the 5970", a $700 card. I would be more inclined to call it two entirely different price points.
And TDP is not determined by the wattage that can be 'potentially' supplied to the GPU. It's the wattage the GPU consumes at max load, its thermal envelope. I'm not sure why so many people have been saying this (because the GTX480 uses six + eight pin PCI-E power connectors it therefore uses the maximum amount of wattage that this configuration can supply, or it therefore uses the same amount of wattage as the HD5970) but this is not necessarily true. I don't know maybe the intent is to mislead those who don't know any better?
The GTX280 used the same connector configuration, and it consumed around 225 watts. The addition of the eight pin was basically meant as a safety against under powering the card. Assuming that the GTX480 consumes 295 watts is an over simplification, and well... misleading.
they love this card so much xfx decided to put triple lifetime warranty on the gtx 470 if u look closely on the box
gbwatson96
Guest
- Jun 29, 2008
- 18
- 0
- 18,510
Again, another over-priced NVidia product coming down the way and we will still be paying $150 for a rehashed 9800GT card. Honestly, Nvidia cards are way to over priced. If ATI had decent drivers, I would buy one in a hot minute. It was nice a few years ago when the BIG TWO fought in out for GPU supremacy. The prices usually got decent after the new releases came out (when the 9 series came out, the 8600GT was a good buy). There are 9800GT video cards out there that are still priced at $140 each (released in 2008)! Everything else since has been a twigged 9800GT of some variety. Why are we still paying over $100 for a GTS 250 card that is nothing more than a modified 9800GTX+ ? In our current economy, making more expensive video cards does not seem very smart. How many people can afford them? Nvidia has a lot of not very good cards in the under $100 price range. You can do a lot better with an ATI product for less money.
- Status
TRENDING THREADS
-
-
News First credible report of RTX 5090 FE with melted connector appears — third-party cable likely cause- Started by Admin
- Replies: 50
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
-
Tested: Intel's Arrow Lake 140T iGPU mostly maintains an edge over AMD's older 880m
- Started by Admin
- Replies: 24
Facebook
Twitter
Instagram