Sign in with
Sign up | Sign in
Your question

New news on GTX 480 & 470

Last response: in Graphics & Displays
Share
February 20, 2010 7:51:48 PM

http://www.semiaccurate.com/2010/02/20/semiaccurate-get...

According to this, things are really not looking good at all.


I don't know how much "Laughabee" hurt the GF100's, but I hope we hear better news soon.

More about : news gtx 480 470

February 20, 2010 8:05:36 PM

Agreed, i really want them to compete with ATI so we can get some good value products :D 
m
0
l
a b Î Nvidia
February 20, 2010 8:07:24 PM

I'm giddy, and I'm sad.

Yay for the PR dog going down.

Bad for no competition.

Hopefully ATI doesn't see this as a reason to stretch the HD5K series into 2011 with only a refresh in Q3.
m
0
l
Related resources
February 20, 2010 8:10:11 PM

Doesn't sound like it's going to happen any time soon. When the BoM for a 480 cost's more then the 5970 retail. And for something to cost more then a product that it can't beat you know heads are going to roll at NVidia.
m
0
l
February 20, 2010 10:14:20 PM

this is nVidia's 2900xt

and the hd 5870 is this gen's 8800gtx
m
0
l
February 20, 2010 10:36:20 PM

shadow187 said:
I'm giddy, and I'm sad.

Yay for the PR dog going down.

Bad for no competition.

Hopefully ATI doesn't see this as a reason to stretch the HD5K series into 2011 with only a refresh in Q3.


The 4-Series came out in 2008, and went into 2009.
A generation of graphics cards is supposed to last 2 years.
m
0
l
February 20, 2010 10:39:51 PM

From what I see here, we see almost nothing. The performance numbers that they quote are far less than credible, and makes little sense given the supposed specs. 512 SPs should atleast come close to double the performance of a GTX 285, even with low clocks. I'm guessing that the GTX 480 will be about 15% faster than the 5870 and about 10-15% slower than the 5970. I also expect the GTX 470 to match the 5870 pretty closely.

But again, my opinion is based on very little fact, just like the article.

That said, Charlie has been pretty accurate.
m
0
l
a b Î Nvidia
February 20, 2010 10:59:15 PM

Another article where he is going over the same points again. And when he hears things he doesn't like, such as points he can't negatively spin anymore, he *SIGHS* and winks , winks that somebody is lying.
Quote:
This speed is the first point the sources conflict on, and it could go either way, both sources were adamant on theirs being the correct final clock. *SIGH*.


Does the same, with pricing/performance, he's doing the worrying for Nvidias accountants, that they are going to somehow have to cheat, to sell a product they engineered. And if there are good reviews, they are 'bought'.
Then he's deflecting the Tesselation performance, its one of the core DX11 additions. Its what probably another area Fermi will perform better in DX11/dx10 games vs ATI.
No matter how Fermi turns out, he's got all bases covered, he's going to tell you he told you so.
m
0
l
a b Î Nvidia
February 20, 2010 11:00:13 PM

AMW, how can a card be 15% faster than the 5870 and 15% slower than the 5970. 5970 is only 30% faster than 5870? :p .

Sabot; HD6k this year -> rumours.
m
0
l
February 20, 2010 11:07:58 PM

shadow187 said:
AMW, how can a card be 15% faster than the 5870 and 15% slower than the 5970. 5970 is only 30% faster than 5870? :p .

Sabot; HD6k this year -> rumours.


Most likely a refresh of the current gen, they won't waste so much R&D for only a year, also they wouldn't have a new process.
GlobalFlo's 28nm process will be ready by the end of 2010 likely but it'll be worse than TSMC's 40nm problems, without a new process they could do an rearrange like
RV670 -> RV770. However RV670 (3870) was RV600 (2900XT) shrunk to 55nm, so it was designed for 80nm. RV770 (4870) took the RV670 & changed it to suit a 55nm process, but Evergreen (RV870) was designed for 40nm, so it's already well very made.
Without a new process they could slightly amp it (5870 -> 5890, 850MHz core to 900-950MHz, 1250MHz RAM to 1300-1450MHz, maybe 1600 SP's to 2000.) but I don't think ATI will call that the 6000 series.
m
0
l
February 20, 2010 11:10:29 PM

notty22 said:
Another article where he is going over the same points again. And when he hears things he doesn't like, such as points he can't negatively spin anymore, he *SIGHS* and winks , winks that somebody is lying.
Quote:
This speed is the first point the sources conflict on, and it could go either way, both sources were adamant on theirs being the correct final clock. *SIGH*.


Does the same, with pricing/performance, he's doing the worrying for Nvidias accountants, that they are going to somehow have to cheat, to sell a product they engineered. And if there are good reviews, they are 'bought'.
Then he's deflecting the Tesselation performance, its one of the core DX11 additions. Its what probably another area Fermi will perform better in DX11/dx10 games vs ATI.
No matter how Fermi turns out, he's got all bases covered, he's going to tell you he told you so.


I agree here, he seemed too vague and full of excuses. That said, I do believe he is accurate, at least, pertaining to the Unigine Heaven benchmark, as I have read in other sources that the benchmark is poor and wont be a good indicator. nVidia should have a slight to moderate advantage in tessellation, but nothing that ground breaking. His cited performance doesn't make sense given the specs. he stated. The clocks are only slightly lower than the GTX 285, and there are more than twice the SPs so the GTX 480 should be ~twice the performance of the GTX 285 which is right between the 5870 and 5970.

shadow187 said:
AMW, how can a card be 15% faster than the 5870 and 15% slower than the 5970. 5970 is only 30% faster than 5870? :p .

Sabot; HD6k this year -> rumours.


Well that is more or less what I meant, right inbetween the two.
m
0
l
February 21, 2010 12:08:45 AM

Maybe Johnnie Walker is clouding my judgment, but the more I read about the Heaven Benchmark score for the GTX480 the more I think the 480 may be a beast.

Look at it like this...

Tessellation is basically a quick rendering of group #'s right. And as of now we have only seen games that use Bitmap, clone and other outdated mapping. But were tessellation shines is collision. a k a light source, light shading, group 3D aspect.

So I guess what I am saying is that either ATI is setting a trap and has on purposely governed their own software to embarrass Nvidia on their PR campaign*. Or Nvidia is about to kick some major ass.


*(Clocked down Tessellation to fool Nvidia)
m
0
l
a b Î Nvidia
February 21, 2010 2:27:25 AM

Tessellation on nVidia is a result of nVid's GPGPU configuration. On any game that doesn't use it, the card isn't good.
m
0
l
February 21, 2010 2:39:48 AM

Tessellation is when a GPU adds details to a mesh/model by using an algorithm on its surfaces (like subsurfs).
m
0
l
February 21, 2010 3:13:44 AM

The article seems to be biased!!!With so much negativity nVidia can never hope to launch them in q2 2010...But according to nVidia CEO Fermi will hit the full stride in q2 2010...which certainly seems contradictory
m
0
l
February 21, 2010 3:17:17 AM

sayantan said:
The article seems to be biased!!!With so much negativity nVidia can never hope to launch them in q2 2010...But according to nVidia CEO Fermi will hit the full stride in q2 2010...which certainly seems contradictory


What are you saying?
m
0
l
a b Î Nvidia
February 21, 2010 3:48:42 AM

Full stride 2011. Shipments of GTX460/480 in 2010.
m
0
l
Anonymous
February 21, 2010 9:07:06 AM

Just want to know how much is charlie paid from ati to write these articles every three day...
m
0
l
a c 172 Î Nvidia
February 21, 2010 9:10:25 AM

sayantan said:
The article seems to be biased!!!With so much negativity nVidia can never hope to launch them in q2 2010...But according to nVidia CEO Fermi will hit the full stride in q2 2010...which certainly seems contradictory

Here's a lovely blog that is to Charlie D what Charlie D is to Nvidia.
m
0
l
February 21, 2010 10:55:31 AM

its not out yet and no credible benchmarks so if we just go looking into history we will prolly see the same thing again - Nvidia = a bit quicker, more expensive. ATi = a bit slower, better bang for the buck.

Thermals, and how much more expensive are another thing all together - might be another dust buster FX5800 ultra...
m
0
l
a b Î Nvidia
February 21, 2010 12:15:52 PM

apache_, from what we know (the process/die size of nVidia), nVid's cards are going to be a lot more expensive.
m
0
l
Anonymous
February 21, 2010 1:09:10 PM

Charlie knows everything about nvidia. Also his source is completely anonymous. What a dumb.
m
0
l
February 21, 2010 3:17:29 PM

sayantan said:
The article seems to be biased!!!With so much negativity nVidia can never hope to launch them in q2 2010...But according to nVidia CEO Fermi will hit the full stride in q2 2010...which certainly seems contradictory

Nvidia is said to be producing only 5,000 to 8,000 of these cards, and it will lose money on each one.
thats full stride? good luck getting one.

The HD 6k is already slated to be a new GPU, most likely ATI's GPGPU. If ATI refreshes the 5000, it will be for the 5890, possibly testing it on the 28nm process, like they did with the 4770. It will be interesting to see who they use for manufacturing, GF is slated to start 28nm in June, while TSMC in q3.
m
0
l
February 21, 2010 3:56:32 PM

ATI RULES!!!!
THE ATI 5970 WILL ALWAYS BE THE BEST!!
m
0
l
February 21, 2010 4:38:03 PM

bhagat666 said:
ATI RULES!!!!
THE ATI 5970 WILL ALWAYS BE THE BEST!!

I love you man
m
0
l
a b Î Nvidia
February 21, 2010 5:01:19 PM

noob2222 said:
Nvidia is said to be producing only 5,000 to 8,000 of these cards, and it will lose money on each one.
thats full stride? good luck getting one.

The HD 6k is already slated to be a new GPU, most likely ATI's GPGPU. If ATI refreshes the 5000, it will be for the 5890, possibly testing it on the 28nm process, like they did with the 4770. It will be interesting to see who they use for manufacturing, GF is slated to start 28nm in June, while TSMC in q3.


Of course you have no links or proof for your Nvidia fables, just slinging it ?
Other than the one site that has gotten you believing that how much profit each card is making Nvidia or ATI for that matter is important to you.
edit: Left AMD out of it.
m
0
l
February 21, 2010 5:43:35 PM

This may be old, but it's new to me.


m
0
l
a b Î Nvidia
February 21, 2010 6:20:08 PM

That's the 'Unigine Heaven' benchmark Charlies was talking about, and notice it's only the Cobblestone Road Sequence, as Charlie mentions reports suggest it's not so go in other areas.

Still early either way, but it's funny that despite having no counter sources everyone wants to simply discount what Charlie says despite his getting the delays and yields correct long before nV finally publicly admitted those issues.

For now I won't go beyond what Charlie essentially says, and what most people were thinking, that the early 'leaked' and PR provided performance figures are simply the best case scenario, and that overall things are looking pretty equal, despite all the talk of Fermi being a monster.

I'll reserve judgment on whether it was worth the wait until the full un-fettered testing is done in a wide range of benchmarks; meaning not the initial PR-controlled launch reviews but the full-on retail reviews with the good & the bad, and perhaps once the DX11 drivers arrive if the other DX11 aspect is true.
m
0
l
February 21, 2010 6:35:01 PM

To be honest, lets assume that that benchmark is correct. So Fermi will match the 5970 in tessellation, that's not all that amazing. I'm not sure why people are saying those results are too good to be true. We know that the Fermi arch. should do a bit better with tessellation than the RV800 arch, so whats the problem?
m
0
l
February 21, 2010 6:36:42 PM

Spoiler
I guess I have a tail, and need to use google more :p 
m
0
l
a b Î Nvidia
February 21, 2010 7:10:18 PM

logitic said:
Spoiler
Do I get a sexy Badge for bursting me tail trying to get info on these cards? :p 

You have a tail?

That makes two furries on THG!
m
0
l
February 21, 2010 7:17:34 PM

Furries, we need some pictures.

Prefurrably on a fermi powered system.
m
0
l
a b Î Nvidia
February 21, 2010 7:40:23 PM

You 'busted yer tail' finding those? One charlie link and the old Unigine results?

May I suggest Google or bookmarks next time to save you the effort. :kaola: 
m
0
l
February 21, 2010 8:27:59 PM

I was hoping the Unigine results picture was new. Didn't know it was old, sorry.
m
0
l
February 21, 2010 10:13:45 PM

notty22 said:
Of course you have no links or proof for your Nvidia fables, just slinging it ?
Other than the one site that has gotten you believing that how much profit each card is making Nvidia or ATI for that matter is important to you.
edit: Left AMD out of it.

Nvidia is said to be producing only 5,000 to 8,000 of these cards, and it will lose money on each one.
thats full stride? good luck getting one.

I figured someone would be smart enough to read the link that started this whole thread, but apparently I was wrong in assuming that would make sense,

click the link in the first post before you try to insult someone.
m
0
l
February 21, 2010 11:20:36 PM

Alright after spending most of the day reading "SemiAccurate" forums I have come to the conclusion that it may be the most biased website on the planet. Each and every time someone would try to take a Nvidia side they were instantly banned. It's felt odd reading all the "Neutral" review talk on that site. To actually see them being super biased.

I am not a Nvidia fanboy, but I am hoping the press release tomorrow destroys that site. What a joke. I can't believe I wasted a day hoping to read some real news.
m
0
l
February 21, 2010 11:28:41 PM

Apparently Nvidia has said to watch their twitter feed on Monday 22nd at 9.00am PST for a 'big announcement'. Everyone's guess is that they're going to be announcing the release date of the GTX 470 and 480. And we can talk all we want about how good or bad they are going to be or how much trouble there has been with yields on the fabrication process etc but how can anyone say anything until we see one in action?
m
0
l
February 21, 2010 11:30:40 PM

Charlie however has been rather accurate lately, he predicted:
RV670 would compete with GTX 260 while being $150 cheaper
Fermi will be delayed to this time this year last year
Ton's of GT212/214/215/216/218/200/b related articles.
m
0
l
February 21, 2010 11:44:12 PM

sabot00 said:
Charlie however has been rather accurate lately, he predicted:
RV670 would compete with GTX 260 while being $150 cheaper
Fermi will be delayed to this time this year last year
Ton's of GT212/214/215/216/218/200/b related articles.


Yes however you must always bear in mind that Charlie seems to rate Nvidia the same way we would rate having a turd on our shoe. Charlie has been accurate yes but he puts NVidia in such a negative light each time he massively exaggerates just how bad these 'truths' actually were. Not wanting to dig ATI, just to give an example, but Charlie in his article linked by the OP slammed NVidia for their poor yields etc and how the 40nm manufacturing process was having such little success. Everyone seems to have forgotten that ATI had the exact same trouble last year and that there is still a mountain of dead 5xxx series wafers at TSMC doing sweet FA. Now do you see Charlie talking about this? No, so take the facts by all means because he's accurate but take the spin he puts on them with a pinch of salt (and by pinch i mean fistful).
m
0
l
February 21, 2010 11:49:26 PM

It is not Charlie's accuracy that I am doubting, it is what he doesn't talk about. I have a feeling he is leaving out parts of the big picture.

For example he says that Fermi will be REALLY HOT, but he may be leaving out that it will have a very efficient cooler to compensate.
m
0
l
February 21, 2010 11:49:53 PM

It's probably very true about nVidia's horrible yields (single digits) because of their experience with 40nm, the size of Fermi, and TSMC's bad fab.
m
0
l
February 21, 2010 11:52:59 PM

Yeah i agree totally, TSMC have apparently got it up to 4% and are 'happy with that'. Rumours are abound that NV is considering the next revision of Fermi already on a 28nm process by Global Foundries. But if we take into consideration just how cavalier NV has been about Fermi i don't think we'll see tapeout (if anything at all) til June next year. There's no doubt this is ATI's golden age.
m
0
l
February 21, 2010 11:56:24 PM

Griffolion said:
Yeah i agree totally, TSMC have apparently got it up to 4% and are 'happy with that'. Rumours are abound that NV is considering the next revision of Fermi already on a 28nm process by Global Foundries. But if we take into consideration just how cavalier NV has been about Fermi i don't think we'll see tapeout (if anything at all) til June next year. There's no doubt this is ATI's golden age.


I've read such rumors, but I am doubtful because of AMD's ties to GF.
m
0
l
February 21, 2010 11:58:29 PM

Yeah that is true, my guess is that they use half GF and half TSMC. Proposterous i know but it would be just like NV to do something like that.
m
0
l
a b Î Nvidia
February 22, 2010 12:08:38 AM

Read harder (and more) homer. ;) 

There's a reason the people taking the 'nVidia side' get banned, because they aren't there to clarify or to provide additional discussion, they're there to protect their favourite company and/or bash Charlie.

Charlie definitely has a hate on for nV, but sofar he's been far more accurate than the others out there who get the leaks. Read one of Theo Valich (BrightSideofNews) of Fuad Abbazovic (Fudzilla) and then if you still think Charlie is alone in this.

The difference being of course that Charlie has actually been pretty good with the nV Fermi rumours since the summer, whereas the other two have been a bit behind the curve with their sources and Fudo essentially repeating nV PR to be different I guess.

BSN has it's good Fermi article where this at least attempts to be balanced (although it still over-estimates nV's yields and underestimates AMD's based on other industry reports);
http://www.brightsideofnews.com/news/2010/1/18/nvidia-g...

Whereas this one calls B3D an AMD site despite it being more of a pure-tech site intolerant of fanbois more than other places, and then goes on to be nV PR by answer the questions;
http://www.brightsideofnews.com/news/2010/2/10/amd-comm...

And he also blames the Oak Ridge cancellation of Fermi cards on some non-existant power requirement, when ORNL is located adjacent to a Hydro Electric damn that they actually run experiments out of, with another only another 10 miles away;
http://www.brightsideofnews.com/news/2010/1/5/the-real-...

The thing with these sites, is in order to be ahead of the other media they need to keep their sources happy, and it doesn't hurt to keep the IHVs happy if you want official 'leaks' given to you for PR purposes. So they usually develop friends & enemies and as such, usually develop a slant. Just trying to figure out who's going to be S3's biatch. :evil: 

I read them all, but I give none of them any weight other than does this or that jive with the situation and other information, and if so what would that mean? But I wouldn't invest money in anything that these rumour sites/people have to say. It's just food for thought and even then it's Junk Food. [:jaydeejohn:7]

Anywhoo, more beer and then back to the game. [:thegreatgrapeape:3]
m
0
l
a b Î Nvidia
February 22, 2010 12:15:01 AM

logitic said:
Alright after spending most of the day reading "SemiAccurate" forums I have come to the conclusion that it may be the most biased website on the planet. Each and every time someone would try to take a Nvidia side they were instantly banned. It's felt odd reading all the "Neutral" review talk on that site. To actually see them being super biased.

I am not a Nvidia fanboy, but I am hoping the press release tomorrow destroys that site. What a joke. I can't believe I wasted a day hoping to read some real news.



I second this. :) 
m
0
l
February 22, 2010 12:19:19 AM

TheGreatGrapeApe said:
The thing with these sites, is in order to be ahead of the other media they need to keep their sources happy, and it doesn't hurt to keep the IHVs happy if you want official 'leaks' given to you for PR purposes. So they usually develop friends & enemies and as such, usually develop a slant. Just trying to figure out who's going to be S3's biatch.[:thegreatgrapeape:3]


So would you say that Charlie has some sources close to AMD that will only give him new information so long as he puts NV in a negative light? Im not having a go but that is what i took from the point you made there.
m
0
l
a b Î Nvidia
February 22, 2010 4:43:24 AM

Griffolion said:
So would you say that Charlie has some sources close to AMD that will only give him new information so long as he puts NV in a negative light? Im not having a go but that is what i took from the point you made there.


No I'm saying he naturally cultivates the sources that are pro-ATi and anti-nV and his articles would tend to accentuate that, and then also people who are probably just friends he acquired while working at the InQ.

Remember he didn't just come out of nowhere with the animosity for nV, alot of it developed after the initial Bump-gate issue on mobile GPUs where nV tried to discredit him instead of coming clean about the issue which later they had to admit in order to cover HP repairs & returns. And even with the initial poor yield and re-spin article (ie for A2 not A3) nV denied that too and said Charlie should check his sources,saying (or lieing as we now know) that yields were fine (which they weren't or else there'd be no A3), and then sending Theo and Fuad out to try and discredit him (although not counter-attacking Charlie by name);
http://www.fudzilla.com/content/view/15689/34/

Just because he has an axe to grind doesn't mean he's wrong. But like I said, I wouldn't put alot of faith in any of them anymore than just talking points. :pfff: 

And now that the game is over and I have time to answer this;
Griffolion said:
Not wanting to dig ATI, just to give an example, but Charlie in his article linked by the OP slammed NVidia for their poor yields etc and how the 40nm manufacturing process was having such little success. Everyone seems to have forgotten that ATI had the exact same trouble last year and that there is still a mountain of dead 5xxx series wafers at TSMC doing sweet FA. Now do you see Charlie talking about this?


Actually he did, and it's not exactly the same either, since ATi had their problems during the height of TSMC's chamber matching issues, where yields were well below 40%, and at that time nV yields were in single digits, now that TSMC's numbers are supposed to be nearer 80% now that the issue is fixed and there is new hardware there then the situation isn't quite the same, hence why there is more focus on the still poor yield of Fermi which is not attributable to the process, unlike that mountain you speak of. The other thing is that nV is already starting from a die per wafer deficit so the yield issue simply makes that worse as well, and thus it's a big issue, especially since the competition has something out there while ATi's yield issues came when nV was ramping down all other production, so the pressure and situation just isn't the same.

m
0
l
February 22, 2010 9:30:00 AM

So are you going to be getting a 470/480? I emailed Scan.co.uk last night and they got back to me this morning saying that they are under a non-disclosure agreement with NV, all i asked was that i be notified as soon as pre-orders became available! :'( 
m
0
l
!