RealityRush
Distinguished
- Oct 14, 2009
- 484
- 0
- 18,790
TheGreatGrapeApe
Champion
- Feb 18, 2003
- 20,043
- 0
- 50,960
Seriously, after all this, and you only have one other 'trusted source' and you can't even understand what is said?
You take this article, which we already discussed a week ago;
That includes this statement;
I asked two people at NVIDIA why Fermi is late; NVIDIA's VP of Product Marketing, Ujesh Desai and NVIDIA's VP of GPU Engineering, Jonah Alben. Ujesh responded: because designing GPUs this big is "****|ng hard".
and turn this quote;
Then timing is just as valid, because while Fermi currently exists on paper, it's not a product yet. Fermi is late. Clock speeds, configurations and price points have yet to be finalized. NVIDIA just recently got working chips back and it's going to be at least two months before I see the first samples. Widespread availability won't be until at least] Q1 2010.
into misreading it ignoring all that is written as if it were confirming Nov availability in your dream world statement;
Talk about taking your only credible source that says Q1 2010 at best, and turning it into something within the next few days/short weeks (how far away do you think Novemeber is? )
Once again, no credible source say anything more than a paper launch before December, and even then it would be a handful of parts to a bunch of trusted reviewers to FUD the waters against any HD5xxx purchase indecision, with NOTHING on store shelves until 2010.
And no credible source says anything about performance, all anyone says is 'sources say Fermi/F100/G300/GT300 is expected to be, will be, must be, needs to be faster than the HD5870; which as you go further you realize the last one is really what they are saying that it needs to be faster than the HD5870 or else it and by extension nV is in trouble with the base it's built it's business on and that which it still needs to finance it's move into other markets.
That white paper tells you very little about how it handles alot of graphical properties and calculations, not the textures even, nor as mentioned aything about the clocks, and whether or not they can do different clock domains still with this design, or if they need all the parts to be at similar or pre-determined ratios for the components frequencies.
There some interesting stuff out there, but it is far from definitive on providing performance information, heck we often don't know that after a few runs of 3Dmark, let alone just a paper outline of the chip diagram.
What you have is hopes, not facts.
edited due to nVidia potty-mouth. 😱
You take this article, which we already discussed a week ago;
That includes this statement;
I asked two people at NVIDIA why Fermi is late; NVIDIA's VP of Product Marketing, Ujesh Desai and NVIDIA's VP of GPU Engineering, Jonah Alben. Ujesh responded: because designing GPUs this big is "****|ng hard".
and turn this quote;
Then timing is just as valid, because while Fermi currently exists on paper, it's not a product yet. Fermi is late. Clock speeds, configurations and price points have yet to be finalized. NVIDIA just recently got working chips back and it's going to be at least two months before I see the first samples. Widespread availability won't be until at least] Q1 2010.
into misreading it ignoring all that is written as if it were confirming Nov availability in your dream world statement;
Talk about taking your only credible source that says Q1 2010 at best, and turning it into something within the next few days/short weeks (how far away do you think Novemeber is? )
Once again, no credible source say anything more than a paper launch before December, and even then it would be a handful of parts to a bunch of trusted reviewers to FUD the waters against any HD5xxx purchase indecision, with NOTHING on store shelves until 2010.
And no credible source says anything about performance, all anyone says is 'sources say Fermi/F100/G300/GT300 is expected to be, will be, must be, needs to be faster than the HD5870; which as you go further you realize the last one is really what they are saying that it needs to be faster than the HD5870 or else it and by extension nV is in trouble with the base it's built it's business on and that which it still needs to finance it's move into other markets.
That white paper tells you very little about how it handles alot of graphical properties and calculations, not the textures even, nor as mentioned aything about the clocks, and whether or not they can do different clock domains still with this design, or if they need all the parts to be at similar or pre-determined ratios for the components frequencies.
There some interesting stuff out there, but it is far from definitive on providing performance information, heck we often don't know that after a few runs of 3Dmark, let alone just a paper outline of the chip diagram.
What you have is hopes, not facts.
edited due to nVidia potty-mouth. 😱
RealityRush
Distinguished
- Oct 14, 2009
- 484
- 0
- 18,790
TheGreatGrapeApe :
***"|ng hard".[/b]
you get...
Talk about taking your only credible source that says Q1 2010 at best, and turning it into something within the next few days/short weeks (how far away do you think Novemeber is? )
Once again, no credible source say anything more than a paper launch before December, and even then it would be a handful of parts to a bunch of trusted reviewers to FUD the waters against any HD5xxx purchase indecision, with NOTHING on store shelves until 2010.
And no credible source says anything about performance, all anyone says is 'sources say Fermi/F100/G300/GT300 is expected to be, will be, must be, needs to be faster than the HD5870; which as you go further you realize the last one is really what they are saying that it needs to be faster than the HD5870 or else it and by extension nV is in trouble with the base it's built it's business on and that which it still needs to finance it's move into other markets.
That white paper tells you very little about how it handles alot of graphical properties and calculations, not the textures even, nor as mentioned aything about the clocks, and whether or not they can do different clock domains still with this design, or if they need all the parts to be at similar or pre-determined ratios for the components frequencies.
There some interesting stuff out there, but it is far from definitive on providing performance information, heck we often don't know that after a few runs of 3Dmark, let alone just a paper outline of the chip diagram.
What you have is hopes, not facts.
you get...
Talk about taking your only credible source that says Q1 2010 at best, and turning it into something within the next few days/short weeks (how far away do you think Novemeber is? )
Once again, no credible source say anything more than a paper launch before December, and even then it would be a handful of parts to a bunch of trusted reviewers to FUD the waters against any HD5xxx purchase indecision, with NOTHING on store shelves until 2010.
And no credible source says anything about performance, all anyone says is 'sources say Fermi/F100/G300/GT300 is expected to be, will be, must be, needs to be faster than the HD5870; which as you go further you realize the last one is really what they are saying that it needs to be faster than the HD5870 or else it and by extension nV is in trouble with the base it's built it's business on and that which it still needs to finance it's move into other markets.
That white paper tells you very little about how it handles alot of graphical properties and calculations, not the textures even, nor as mentioned aything about the clocks, and whether or not they can do different clock domains still with this design, or if they need all the parts to be at similar or pre-determined ratios for the components frequencies.
There some interesting stuff out there, but it is far from definitive on providing performance information, heck we often don't know that after a few runs of 3Dmark, let alone just a paper outline of the chip diagram.
What you have is hopes, not facts.
I have a lot of hopes, not gonna lie there, hopefully they aren't dashed.
But the Anandtech article says MAINSTREAM cards are scheduled for Q1 2010, the high end cards would logically be released before that... it's just a matter of how much earlier, and hopefully that "earlier" is Nov. 2009.
wiinippongamer
Distinguished
- Sep 9, 2009
- 1,927
- 0
- 20,160
hope they did, I'm waiting till january for upgrading just for gt300; and i'm not planning to get ati because of their annoying faulty drivers.
daedalus685
Distinguished
- Nov 11, 2008
- 1,558
- 1
- 19,810
Uhg...
ATI drivers have been no worse than Nvidia since the 8000 series... Both have issues, and both work fine 99% of the time.
That kind of ridiculous fanboy jargon doesn't help anyone..
RealityRush
Distinguished
- Oct 14, 2009
- 484
- 0
- 18,790
Ehhhh I'd say ATI had problems with drivers later than that... granted the later 4800's and the 5800's seem to not have any driver problems, so I don't think that's much of an issue anymore.
The only real reason to wait for the GT300's at this point is them outperforming anything on the market with DX11 currently.
I'm hoping that's the case, cause if it's not........ I'll stab something for making me wait that long.
smithereen
Distinguished
- Oct 4, 2008
- 1,088
- 0
- 19,310
- Jul 24, 2006
- 22,978
- 312
- 57,490
wiinippongamer
Distinguished
- Sep 9, 2009
- 1,927
- 0
- 20,160
im saying it from my own experience, i've always used nvidia except from a x1550 years ago, it crashed on almost any game and screen flicked,i waited a few months for a new driver to come out, installed it, and still, nothing. i googled like crazy and found a lot of people having the same problems.
i just decided to give up and bought the expensiver, still faster but way much more stable 6800gt
btw i had XP since it came out and i'm still using it so i never had any kind of problems with nvidia.
for what i've read now ati has LESS problems, but they still have it, plus it seem that since the 3800 the powerplay has been causing slowdowns.
JAYDEEJOHN
Champion
- Aug 23, 2006
- 19,840
- 0
- 50,860
The old "fall back on the faulty driver" baloney.
Every once in awhile, I go into a nVidia driver problem thread and try to help. and pccasionally, let others already subscribed to that thread that driver issues arent exclusively ATI's, as some here know.
This is an old and tired excuse to "think" that nVidia is better, which it truly isnt.
Divide the Vista numbers, and youll see that nVidias drivers were 40% worse than ATIs drivers, per market share/user.
Look, if you need something to convince yourself that the G300 will kick some major booty, then do as the OP has done, and search the internet for info, and dont be stupid and play down ATI, its drivers or any other excuse.
nVidias in a tough situation, and anyone that doesnt acknowledge that just isnt getting it.
They dont have any DX10.1 cards that are mid or higher, no DX11 cards, their old cards are priced too high and EOL, and theyre hoping for nVidia fans to wait.
Nothing wrong with waiting, but there is when you have to slam the competition, to make the wait worthwhile, while the competitions users are making their gaming worthwhile on their new DX11 cards
Every once in awhile, I go into a nVidia driver problem thread and try to help. and pccasionally, let others already subscribed to that thread that driver issues arent exclusively ATI's, as some here know.
This is an old and tired excuse to "think" that nVidia is better, which it truly isnt.
Divide the Vista numbers, and youll see that nVidias drivers were 40% worse than ATIs drivers, per market share/user.
Look, if you need something to convince yourself that the G300 will kick some major booty, then do as the OP has done, and search the internet for info, and dont be stupid and play down ATI, its drivers or any other excuse.
nVidias in a tough situation, and anyone that doesnt acknowledge that just isnt getting it.
They dont have any DX10.1 cards that are mid or higher, no DX11 cards, their old cards are priced too high and EOL, and theyre hoping for nVidia fans to wait.
Nothing wrong with waiting, but there is when you have to slam the competition, to make the wait worthwhile, while the competitions users are making their gaming worthwhile on their new DX11 cards
jennyh
Splendid
- Mar 3, 2009
- 4,292
- 0
- 22,860
http://www.tomshardware.co.uk/geforce-gt-220,review-31703-13.html
Scroll down to see those awesome nvidia drivers in action with dx10.1
Scroll down to see those awesome nvidia drivers in action with dx10.1
wiinippongamer
Distinguished
- Sep 9, 2009
- 1,927
- 0
- 20,160
JAYDEEJOHN
Champion
- Aug 23, 2006
- 19,840
- 0
- 50,860
My last statement was a lil harsh, about people gaming on their 5xxx series, but true none the less, and I thought about retracting it, but yes, I favor the red team, not so much as Ive never owned an nVidia card before, had several, and enjoyed them.
But, facts are facts, my experience with nVidia and ATI for what its worth, never had a major problem, ever.
All things equal, I buy ATI, when it isnt, I buy nVidia or ATI when it offers more perf or price/perf in the area Im looking.
When I say nVidia is hurting right now, its true, not a malicious statement whatsoever.
They NEED G300 across their range, cause their old cards cant compete, and they havnt lowered prices due to their pricing schemes, which guarantees the seller a fixed sales price, or nVidia reimburses them.
They NEED G300 to be a killer gpgpu, and they need to clear stock as well.
Theyve released 3 cards since January? , while ATI has had 5, 4 in the new DX model, and the 3 we wont see as enthusiasts, as theyere entry level type cards, and by the time G300 does come out, and maybe 1 lil brother with it, ATI will have their entire market covered.
But, facts are facts, my experience with nVidia and ATI for what its worth, never had a major problem, ever.
All things equal, I buy ATI, when it isnt, I buy nVidia or ATI when it offers more perf or price/perf in the area Im looking.
When I say nVidia is hurting right now, its true, not a malicious statement whatsoever.
They NEED G300 across their range, cause their old cards cant compete, and they havnt lowered prices due to their pricing schemes, which guarantees the seller a fixed sales price, or nVidia reimburses them.
They NEED G300 to be a killer gpgpu, and they need to clear stock as well.
Theyve released 3 cards since January? , while ATI has had 5, 4 in the new DX model, and the 3 we wont see as enthusiasts, as theyere entry level type cards, and by the time G300 does come out, and maybe 1 lil brother with it, ATI will have their entire market covered.
- Jul 24, 2006
- 22,978
- 312
- 57,490
Here is my 2cp worth.
If I recall when Nvidia released thier gt200s, there was speculation due to the size of the chip and the cost to manufacture and sell it, Nvidia could not afford to take a price cut on this line. Today that is showing more than ever since they haven't lowered thier prices but instead are cutting 55nm production lines.
If GT300 is not faster than Radeon 5800, Nvidia cannot afford to even sell the product. 50% more transistor count = 50% more production cost. They have to be able to charge a premium in order to stay afloat, and if this card underperforms, they can't sell it at all. That is why the GT300 isn't about how fast it is, its about how long it takes Nvidia to make it faster than AMD's competition. This is also why the first GT300s will be Tesla GPGPUs. They will be selling for $1500+, no matter what, thats a profit.
If I recall when Nvidia released thier gt200s, there was speculation due to the size of the chip and the cost to manufacture and sell it, Nvidia could not afford to take a price cut on this line. Today that is showing more than ever since they haven't lowered thier prices but instead are cutting 55nm production lines.
If GT300 is not faster than Radeon 5800, Nvidia cannot afford to even sell the product. 50% more transistor count = 50% more production cost. They have to be able to charge a premium in order to stay afloat, and if this card underperforms, they can't sell it at all. That is why the GT300 isn't about how fast it is, its about how long it takes Nvidia to make it faster than AMD's competition. This is also why the first GT300s will be Tesla GPGPUs. They will be selling for $1500+, no matter what, thats a profit.
smithereen
Distinguished
- Oct 4, 2008
- 1,088
- 0
- 19,310
- Jul 24, 2006
- 22,978
- 312
- 57,490
TheGreatGrapeApe
Champion
- Feb 18, 2003
- 20,043
- 0
- 50,960
I think that's once again taking nV at their word, and recently their word has been mud. Let me fix that for you
;- But that's mainstream, so high end should be earlier. When and by how much? Nobody knows yet. And will depend on nV first getting the F100/G300 out to begin with, and even that may be Q1 at the current rate. Hopefully it's sooner rather than later, but no one knows for sure when they will really arrive (in quantity greater than double digits), likely even nV doesn't know right now. -
Sofar the timeline still relies on alot of things that need to succeed, alot of which would still not be finalised yet (like final PCB design testing, clock speed finalizartion, driver alpha testing, etc).
At most we are hearing 'Best Case Scenario' dates, it's not like they are way ahead of schedule and can launch any second, they are going to be slaves to each separate step in the process, unlike some of their prior parts where they could just pick an arbitrary date long before it nears.
My guess on timelines;
F100/G300/GT300 Announce with a tiny handful of card for late November - mid December at the latest (anything later won't affect ATi sales and just confirms their 'paper launch' reputation), with availability in mid - Q1.
Mainstream part, paper launch details at the same time as the high end, with handful of availability in early Q1 2010 (once again for reviewers mainly), but likely being smart about not paper launching again so they can pretend it's the same as everyone else's days before availability launches, and then wider availability late Q1 likely even early Q2.
nV has not been good at execution as of late, so I wouldn't think everything will go perfectly for them, and that's the only way those rumoured timelines hit when nV themselves admitted a late first batch (and even the chip said week 39, remember?).
I would say take any date nV gives you and then... add 2-8 weeks for shipping & handling. :lol:
TRENDING THREADS
-
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
-
-
Question would like to know if the products i chose is compatible or not
- Started by undervoltt1214
- Replies: 4
-
-
News Leaked RTX 5060 Ti 16GB benchmarks show a 20% uplift over the 4060 Ti 16GB- Started by Admin
- Replies: 15
Facebook
Twitter
Instagram