AMD = performance
AMD's R600 leads Nvidia by about 20% even with less and slower memory. The R600 is 29.6% faster doing Call of Duty 2, 21% faster on Oblivion, 16.3% on 3DMark06, 29% faster on Cadalyst C2006, 29% faster on Autodesk 3ds Max v8 OpenGL, 41.1% faster on Autodesk 3ds Max v8 D3D, 26.4% faster on Catia 02, 58% faster on Maya 02. It was a total slaughter.
Those tests with small differences are clearly sign of CPU bottleneck. Wait till you see it on K10.
Just to remind you, the 2.5GHZ K10 is 33% to 66% faster than Intel's 3GHZ quad core.
posted by Sharikou, Ph. D. @ 4/24/2007 12:19:00 PM
58 Comments:
The only problem is Linux drivers, right ?
As Ho Ho pointed out the HL test were dodgy which makes you wonder about how valid the other tests are.
Also the 2nd set of tests (from the dailytech website) were against a Quadro which is old technology
Looks like AMD needs something.
From Rueters:
"Intel's share of the $30 billion market for x86 processors that power most personal computers was 80.5 percent in the first quarter, according to Mercury Research, a market tracking firm whose data are closely watched by the industry.
That represented a gain of more than 6 percentage points from the 74.4 percent Intel had in the fourth quarter.
Intel's gain came at the expense of Advanced Micro Devices Inc., which saw its share fall to less than 20 percent for the first time since 2005, Mercury said.
"
Hector Ruiz is going to prison.
A week ago he knew they were doing a private debt placement, but yet did not mention it when asked during the conference call.
AMD senior executives are guilty of securities fraud.
I again just can't believe these ridicules Intels fanboys, they really just don't understand!
These are not Intel fanboys , I have reasons to belive that these are Intel paid button pushers repeating the same stuff, which is non-relevent ,over and over again on this blog.
Just look at bubba's posts above.
All users like bubba who keep saying the same sh*t over keep their profiles hidden and are all registered in the past 2-3 months.
Dell says only Intel systems are next generation.
AMD junk doesn't get new features.
http://www.dell.com/content/topics/global.aspx/corp/pressoffice/en/2007/2007_04_24_rr_000?c=us&l=en&s=corp
AMD Vice President John Byrne at the big press junket today:
"We've had our arse kicked."
"30 per cent of AMD products being sold at less than $49 and three per cent at the more than $200 end of the scale."
How good are those AMD products again? Not very if what they are worth is any measure.
AMD debt downgraded to "junk" grade, just like it's products.
Wow this is easy. I guess I'll stop for now, there will be more bad news for AMD fanboys tomorrow.
http://yahoo.reuters.com/news/articlehybrid.aspx?storyID=urn:newsml:reuters.com:20070423:MTFH16761_2007-04-23_22-26-00_N23338706&type=comktNews&rpc=44
bubba and other button pushers:
You can try and spin it however you want, the fact remains that fewer and fewer people are buying into you propoganda. Time to change tactics, you and your company truly are starting to sound very desparate. While crashing blogs and spreading anti-AMD/pro-intel might have worked as a strategy to sway public opinion for the short term, users will dump your products like hot potatoes when they see what true innovation can do.
And I truly question your antics about trying to refute real benchmarks offered by AMD, when you fellas attempt to insult the masses intelligence with things like Penryn being %211 faster in DivX. If that wasn't a blatant attempt at pulling the wool over the sheep's eyes, I don't know what is.
It's time to take your marketing campaign in a new direction if you have any hope of survival. How long do you think people will stand for it? I'll tell ya, about another month.
Lets look at them AMD cumm lappers
"Everything is going perfectly"
Another cumm lapper "I again just can't believe these ridicules Intels fanboys, they really just don't understand!"
And this last cumm lapper " You can try and spin it however you want, the fact remains that fewer and fewer people are buying into you propoganda. Time to change tactics, you and your company truly are starting to sound very desparate. While crashing blogs and spreading anti-AMD/pro-intel might have worked as a strategy to sway public opinion for the short term
Yup we push the same buttons don't know $hit, spread propoganda.
THis ain't propoganda, this is facts dude. If you can't comprehend you need to change your cumm filled diet.
AMD on track to losing billions, running out of cash, stopping investing in manufacturing, losing Market Share, one year+ late at 65nm, going to be even later on 45nm. Barcelona samples should be everywhere for everone to see how great it is, and where is it? Still hiding behind vague annocements. Leecooper got it right "Everything is going perfectly"
From "Ph"ony: AMD=Performance?
More like AMD+ATI = BK
ATI graphics chips have no correlation to AMD Barcebologna performance.
The latest graphics were designed and Taped Out long before AMD had any opportunity to ruin them.
Just the same ATI team joined AMD to late to influence Barcebologna. It was spoiled meat before ATI could help fix it.
More "Ph"ony logic from the "Ph"ony "D"octorate
I would like to reference another one of the "Ph"onys famous prediction here in this blog
Thru his "Ph"ony logic he claimed AMD would be at 55% market share.
Well the numbers are in Sharikou, you were a bit of with your "Ph"ony prediction. AMD is lowing MS faster then anyone thougt going down to 80%. I expect it to drop another 3-5% this quarter.
Barcebologna is too little to late.
You AMD cumm feeders let me refresh your memory and the flawed logic from your "Ph"ony "D"octorate
http://sharikou.blogspot.com/2006/03/amd-poised-to-exit-2006-with-55-market.html
Intel Wipes Out AMD's 2006 Marketshare Gains in One Quarter
Intel erased all of AMD's 2006 gains and pushed the company back under the 20 percent threshold. Analysts for JP Morgan reported that "Intel's superior products and aggressive pricing took their toll on AMD. We expect AMD to lose additional share during (the second quarter)."
No one wants AMD's old 90nm ancient junk when they can pick up a Core 2. I predict AMD will exit 2007 with less than 10% of the marketshare.
AMD BK Q2'08
No one wants AMD's old 90nm ancient junk when they can pick up a Core 2. I predict AMD will exit 2007 with less than 10% of the marketshare.
AMD BK Q2'08
I agree wholeheartedly. AMD is finished. AMD is going further into debt to finance it's operations. Billions of debt, only six months of cash left.
I think that netrama and co. are not actually AMD fanboys, but are rather paid by AMD to try and prove that AMD actually has supporters.
Sharikou is a fool. Look at the link he linked to. Who cares about 3DMark? We're looking at the REAL game performance. The R600 does not make that strong a performance gain over Nvidia's kit to warrant that six month delay. Nvidia has new parts on the way that will beat R600 with ease. They've had six months to prepare the counter attack. Why isn't the XTX benchmarked? The XT is a little quicker than the GTS. The GTX will be easily faster than the XT. The XTX will be needed to compete with the 8800 GTX. Even more so, since the 8800 Ultra will be out shortly.
Sharikou is a fool alright. I just re read his post. The 2900XT has 512mb of 1.65ghz memory, connected over a 512bit bus. The 8800 GTS has 640mb of 1.6Ghz memory, running over a 320bit bus.
The Ati card has a memory bandwith advantage, and the Nvidia card has an overall memory capacity advantage.
An extra 128mb of memory won't make as much difference as the extra memory bandwidth the Ati card has.
Hector Ruiz should be fired. In the last year, Hector Ruiz has done the following:
-Burdened AMD will billions of dollars in debt due to buying Ati
-Wiping out shareholder value. AMD was worth 20bn last year, now it's worth 7bn. 13bn gone just like that.
-Reduced DAAMIT's combined revenue from 1.92B to 1.23B, a 37% drop for Q1.
-Reduced DAAMIT's combined profit from just short of 200m to a loss of 600m in the Q1 alone.
-AMD's market share has dropped to 19%, down more than 6% in Q1 alone. Expect it to drop further throughout the year.
The blame for all of this falls solely on the shoulders of Hector Ruiz.
If AMD is acquired, I wouldn't be suprised to see the whole management team fired. The GPU company would be sold to someone else, or spun off from AMD. With the management gone they could bring in some fresh blood and return AMD to glorious company it once was.
Ha, good to see AMD use a Core 2 Extreme to benchmark R600!
wakakakkaa... funny funny..
congratulation sharikou.. your site is really really attracts all the intel fanboys attention.. One by one of the intelers might get heart attack while reading your post.. you know... ekekeke
Here's an interesting fact:
Anyone over the age of 12 that has ever posted here made more money last year than AMD has made in it's entire 38 year existance.
Oh great. He posted the Quadro benchmarks. The Quadro card benchmarked against is based on the G71 core as was stated earlier. The Quadro FX 5600 and 4600 are the current Quadro cards based on G80. You can't compare an unreleased card to a card that's a generation old. That's just plain stupid.
Also benchmarking against an 8800 GTS was pretty misleading, the 8800 GTX is Nvidia's top of the line card, and it's a good deal faster than the GTS.
Since the 8800 Ultra will be going against the R600 XTX, it seems only fair to compare the XT to the GTX. The GTS will fall in price a bit, and would compete with the "XL" version of R600, should Ati release one.
As a final reccomendation, you should remove the "20% even with less and slower memory" line. That's misleading. The Ati card has faster memory. It has a 512bit memory bus vs. 320bit on the Nvidia. The GTS has 64/GB of memory bandwith. The Ati card has (according to my awful mathematical abilites!) 105.6/GB of memory bandiwth. That's 65% more! The Ati card has a massive memory bandwith advantage, that more than makes up for the 128mb more memory that the GTS has.
Everyone just has to love how Sharikou can so blatantly distort the truth. Does R600 return Ati into direct competition with Nvidia? Definitely. Is it the clean kill Sharikou proposes? Not at all.
Re: Those STUPID numbers that Sharikou keeps citing about the K10, (fudzilla is even less reliable as a news source than the inquirer!)
AMD may well be as fast as they claim at the same clockspeed. If we add up the performance gains AMD believe to be true, 50% in FP, 20% in Integer at the same frequency. We arrive at the general conclusion that K10 will be ~35% more efficient than Clovertown, at the same frequency.
But, Penryn will be fighting Barcelona, not Clovertown alone. Intel has shown off a few benchmarks, I think it's fair to say that an average IPC increase of 10% is about accurate. So if we accept AMD's claims and valid and true, there we have it: Barcelona will be 25% faster at the same frequency than Penryn.
So a 2.5Ghz (the fastest at launch) Barecelona will be (roughly) as fast as a 3Ghz Penryn, and would be slightly faster (~10%) than a 3Ghz Clovertown.
These numbers are by no means concrete, just an estimate, but they make a lot more sense than the numbers Sharikou pulls from his arse!
Will Barcelona close the gap, and ensure that AMD and Intel are about equal on performance in servers? Yes. Will it be 200% faster like Sharikou claims? Not a chance.
Fudzilla explains that the R600 XTX has been delayed again, I don't know if this is true or not, but it seems to be a pretty good explanation for why Ati isn't showing off the XTX.
R600XTX, the 1024 MB GDDR 4 card, has been pushed to the next quarter. This is just one in a series of ATI's failures, but of course DAAMIT will call this a strategic decision. R600XTX won't see the face of retail / etail stores till Q3 2007.
We know that something went wrong with the samples and that the there were some severe performance problems with the latest batch of GDDR 4 cards. Basically 512 MB DDR3 version with 1600 MHz memory was beating the GDDR4 version with 1024 MB of GDDR 4 memory clocked at 2200 MHz.
That cannot be good as the GDDR4 card is much more expensive to build or sell. The worst part is that there is no any big performance difference between 512 GDDR 3 card and 1024 MB GDDR. The trouble is that 1 GB of GDDR 4 at 2200 MHz costs you a lot of money, so ATI at least wants to save some money. It will launch this card after Geforce 8800 Ultra but remember, don’t expect too much.
So as it was stated, the 8800 GTX will compete against the R600 XT and not the 8800 GTS that they're comparing it against now. I'd expect Nvidia to drop prices on the GTX to match the prices Ati will set for the R600 XT. The 8800 GTX will be slightly faster than the R600 XT, but it won't be a huge victory. It seems that the 8800 Ultra will take up the place the GTX occupies now as the highest end card. It will be unmatched until the R600 XTX eventually find it's way out into the market in Q3.
One thing's clear though: The 512bit memory bus on the R600 gives Ati a huge memory bandwith advantage even against the GTX, but memory bandwith alone isn't enough to determine performance. The GTX has 50% more memory than the R600 XT, and that would certainly come into effect at high resolutions.
People, look at this guys, they are comming in hordes.
:)hm, just unbelievable!
And your engrish is pheonominal...I salute you....
AMD/ATi should have the performance lead after being delayed so long, it would not be prudent to release it with similar performance to nvidia who has had a card out 6-7 months already. I would guess the R600 will really shine at high resolutions but to be honest both nvidia and AMD/ATI cards already kill most 3d games and it will be how they handle DX10 that needs to be analysed to see who is best overall.
nVidia are tough customers so it will be an interesting battle.
lol Out of the woodwork as they say! Riding in in a blaze of glory, guns cocked (or should I say cocks cocked), yelling yippee kiya mother**cker! Hahaha Just adds credence to you fellas being paid shills. You do realize you are doing more harm than good, no? Well, you probably don't quite get it i'm sure as it seems to me a recent study found Intel users to have a lower IQ than AMD users. Anyway, keep it up and thanks for the boost to AMD's image! :)
Sharikou, any indication of unusual high Intel IP addresses stemming from Intel as on Scientia's blog?
This comment has been removed by the author.
This comment has been removed by the author.
LG said...
...Anyway, keep it up and thanks for the boost to AMD's image! :)
good one! I like this. Sharikou, keep up the good work portraying good image for AMD and showing off high IQ of the AMD fans through your blog. Your blog is a reflection of the AMD fans' IQ. You are the inspiration for a lot of AMD fans and I believe LG would love to have the IQ as high as yours.
LG, read more this blog, think more like sharikou, and you would be successful. God bless you. AMEN
Does Intel customers satisfied with their core 2 duo processors?
http://www.vector64.com/Conroe/d4600y.html
http://forumz.tomshardware.com/hardware/Core-Duo-running-slow-ftopict209110.html
http://forums.pcpitstop.com/index.php?showtopic=135019
http://forumz.tomshardware.com/hardware/P5B-Problems-Conroe-Build-Help-ftopict212629.html
http://au.answers.yahoo.com/question/index?qid=20070106115238AAlYwHC
and more...
scientia is totally right..
http://scientiasblog.blogspot.com/
i sure hope AMD can make a lot of money with behind-closed-doors benchmarks because Q2 is surely on a roll with AMD at a downward spiral 18% market share, with half a Billion $ quarterly losses with a $900M product inventory.
I hear reporters are buying benchmark stories and AMD presentation slides at $1.50 a pop.
Now if AMD can only make 300 million more benchmarks to sell to news reporters they would just break even. Heck, they don't really need to ship out anything new like they're already doing today.
Or they could also divest in the tree planting business... what? they just did with Dell?
what a loser company.
Damn.. Intel customers need to be more careful. They have to ensure that their pcs are switched off especially while leaving their houses..
otherwise..
http://consumerist.com/consumer/dell/dell-laptop-burned-down-my-house-235167.php
Does Intel customers satisfied with their core 2 duo processors?
Of course they are. 6% of the market decided that Core 2 Duo offers excellent performance superior to that of AMD's products and switched to Intel accordingly. Now, you'll probably try and argue that plenty, but the numbers don't lie. Now basically all of Intel's desktop and mobile processors that are popping out of Intel's fabs are based on the Core Micro-Architecture. (Not counting servers, since Tulsa is a netburst CPU)
scientia is totally right..
yeah, scientia also said AMD will end 2007 with 30% market share. He said C2D will have no impact to AMD.
I could go back and give you a long list of things he gotten wrong but i really can't be bothered.
There is just too many horrible news and so many things going wrong for AMD that I feel pointing them out would just be tactless. But I sure do hope you all enjoy being in the back seat again waiting for the day AMD "delivers". Hoping and praying for the day to arrive is something I'm sure you're all familiar with.
Oh, this is even better.
Damn.. Intel customers need to be more careful. They have to ensure that their pcs are switched off especially while leaving their houses..
Could you explain why it's Intel's fault that Dell used faulty Sony batteries in their laptops? Sony has recalled all the batteries, at a cost of hundreds of millions of dollars. I think it's safe to say that they've taken responsibility for this, by recalling the affected batteries.
"Fortunately, because of the ambitious program Intel put into place at the outset, Intel has the ability to remediate potential losses in an individual's e-mails from a multitude of alternative sources, and has a sound basis to believe that ultimately nothing of any genuine significance will prove to have been lost," Intel said in the 42-page report.
LIES LIES LIES!!!
"Could you explain why it's Intel's fault that Dell used faulty Sony batteries in their laptops?"
Can you explain what are these two guys talking about?
ipinkus said:
I totally believe this. Twice now my Dell Inspiron 6400 power adapter tried to burn down my house... The label bubbled and burned off it the first time I noticed it overheating..
johnnygee73 said..
I did buy a Zalman 100% copper heatsink (w/fan included) and hooked it all up and made sure all fans were turning in the right directions. Still getting an overheating issue...The CPU temp spikes to sometimes 200 degrees celsius for a split second.
"Of course they are. 6% of the market decided that Core 2 Duo offers excellent performance superior to that of AMD's products and switched to Intel accordingly"
And 94% of the customers claimed that their core 2 duo processor damn hot.
pezal, in a remarkable leap of faith, reported:
I totally believe this. Twice now my Dell Inspiron 6400 power adapter tried to burn down my house... The label bubbled and burned off it the first time I noticed it overheating..
So, according to your elementary logic, an overheating transformer core, connected to a battery, which feeds the power supply, which feeds the voltage regulator on the motherboard (amongst other things), which then powers a CPU... is overheating because of the CPU? What an amazing leap of faith and a complete lack of comprehension of causality. Couldn't be that the power adapter was a cheap piece of crap that cost Dell $1 less than a better adaptor, allowing Dell to pocket another $1? Nooooo, must be the CPU. Don't try to use that adapter on an AMD CPU-based laptop. You might find yourself disappointed with the results (hint: it's not the CPU causing the problem).
Sheesh. Critical thinking skills are clearly not required for posting.
dr,yield,phd,mba in a remarkable leap of faith, reported:
Based on my research and findings, I totally believe that Intel CPUs does not need Heat Sink and fan anymore.. wakakaka.. so that it can be quickly and easily melted and burn the laptop together with the house of the laptop owner.. ;-)
Whoops, another downgrade for AMD, this time from Thompson.
Who the hell is going to buy 6% notes from this company?
And 94% of the customers claimed that their core 2 duo processor damn hot.
Nope. Of the remaining 94%, 75% of those people use Intel CPUs. 81% of computers use Intel processors. Intel CPUs are a "smarter choice". AMD's market share has dropped from 25.5% to 19.5%.
Do you honestly know anything about CPUs? If a CPU was running at 200C it would blow up. No CPU can withstand that heat. Clearly, the program used to report the temperature, or the temperature sensor itself were wrong.
I have a Core 2 Duo and the temperature never exceeds 45C.
So much for R600.
Not even out yet and can't beat a six month old Nvidia card.
http://www.dailytech.com/Overclocking+the+R600/article7044.htm
"Do you honestly know anything about CPUs? If a CPU was running at 200C it would blow up. No CPU can withstand that heat."
Why ask me? You should ask johnnygee73 why is that happen.. here is the link..
http://au.answers.yahoo.com/question/index?qid=20070106115238AAlYwHC
My guess is that Johnnydumbshit, along with most of you guys supporting Sharidouche, hasn't the slightest clue what he's doing, and shouldn't be allowed to even look at the inside of a PC much less put his hands in it...
giant
"If a CPU was running at 200C it would blow up."
I'm not sure about CPUs but old Voodoo GPUs (older name: vpu) used to run at >130C without problems. Still not quite 200C but considerably more than most CPUs.
From later GPUs I know 6600GT often went >120C and mostly worked quite nicely. Also I've personally seen a 3GHz Prescott running under full load at 100C for 24h without any stability problems. Though that was with loose heat sink and CPU was throttling.
The temperature spikes and voltages they report there are a bug of motherboard software, not a problem with CPU.
Anyone catch the news about Acer doing a battery (Sony) recall?
Guess who's processors are in the machines... Intel, yes... AMD... yes they are.
Aspire 9300.
I did not look at all of the effected models, but needless to say it only takes one!
How will Sharikou spin this and say it was AMD's processors, oh wait a minute, he won't... it must be the batteries...LMFAO!!!
This comment has been removed by the author.
Not even out yet and can't beat a six month old Nvidia card.
Yes. I stated that earlier. Sharikou posts results comparing the R600 to Nvidia's second fastest card. But the GTX is much faster than the GTS, and is more than enough to frag R600. Even worse, the Quadro results are from a product based on G71.
Well read it a weep intel fanboys.
I hooked a watt meter to some intel and amd platforms and I think the results speak for themselves.
A e6600 c2d on a intel 945 platform draws 78 watts at idle and 94 watts while surfing the internet and 107 watts when both cores are running at 100% load.
A AMD AM2-4800 brisbane on a standard amd platform motherboard draws 46 watts at idle, and and 50 watts while surfing the internet and 104 watts when both cores are running at 100% cpu load.
So if your one of those people who leave your computer on 24/7 and you have a c2d you are using twice the power of my AMD.
If you have a c2d and you are reading this blog then you are burning twice the power of my amd 4800 brisbane.
If intel cant produce energy saving cpus and motherboards you should only buy AMD.
Save the planet, stop using those room heating c2d chips and platforms.
oneexpert
What you fail to have included in your test is that the E6600 doesn't compete against the X2 4800+.
The E6600 really competes against the FX62 or the X2 6000+, so your numbers, while showing lower power consumption for AMD, they are not quite representative of the performance or the performance per watt difference between AMD and Intel.
oneexpert
"50 watts while surfing the internet and 104 watts when both cores are running at 100% cpu load"
So running the CPU at full load increased power usage on AMD by 54W wheras on Intel it only increased by 13W. Are you sure you measured it correctly?
Here are the numbers from a reputable site: http://techreport.com/reviews/2007q2/core2-qx6800/index.x?pg=13
E6600 uses less power under load than all AMD CPUs except 3600+ (slowest X2)
Hey Sharikou, here's a link for you:
http://www.dailytech.com/article.aspx?newsid=7052
R600 consecutively fragged by Nvidia's current 8800 GTX even before the official launch!
What do you have to say about that? Not even the R600 XTX can stand up to the mighty Geforce 8800 GTX.
giant
"Not even the R600 XTX can stand up to the mighty Geforce 8800 GTX."
I'm quite sure XTX can fight against GTX and they should end up even. Ultra is a whole new story, though.
It is kind of funny how R600 has almost all the numbers way better than NV but still doesn't have enough performance:
320* SPUs vs 128
512bit vs 384
1G vs 768
80/65nm vs 90nm
2.2Ghz vs 1.8GHz
Only thing in NV advantage is its shader speed: 1.35GHz vs 0.8GHz
*) It is actually a bit more difficult. R600 uses same kind of shader architecture as most older GPUs where you have 4-way SIMD and one scalar unit per ALU. In total R600 has 64 such ALUs and each of them is capable of up to 5 FP calculations. G80 has 128 scalar units with no SIMD each doing a single FP calculation.
I'm quite sure XTX can fight against GTX and they should end up even. Ultra is a whole new story, though.
Incorrect. See the link I provided earlier. The XTX is totally fragged by 8800 GTX. AMD is embarrasing. Not only is their product six months late but it also has inferior performance to Nvidia's existing cards.
I know it got fragged. I also know that they got results on XTX that showed it to be slower than XT. It is not exactly reliable source of information. My guess is they didn't use correct drivers or that drivers are simply faulty.
NV should get its 65nm high-end out in H2 this year, a similar upgrade that was with 7800->7900. What will AMD have to put against that one? 2900xtx comes out some time in June being several months late and will be replaced by something in a couple of months? Kind of sounds similar to what happened with R520.
In any case, a lot of money will be wasted or AMD won't be able to earn profits from highest-end. Considering that NV had bigger revenue from selling its workstation GPUs than whole AMD GPU lineup I'd say there is a lot to loose when you can't make a decent workstation GPU.
More people are speculating that AMD will BK by Q4'07 if they aren't acquired by someone:-
I hate private equity, but AMD won't last the year unless Silver Lake or Francisco Partners jumps in to save them
They suggest bringing back Jerry Sanders and kicking Hector Ruiz out. That's a solid idea. Ruiz and his executives squandered the time they had a huge performance advantage over Intel. Now Intel is fragging them on all fronts. What a waste.
http://www.eetimes.com/blog/news/archives/2007/04/maybe_jerry_san.html;jsessionid=MALU3ODXRVIHYQSNDLPCKH0CJUNN2JVN?loc=semiconductors
http://www.dailytech.com/article.aspx?newsid=7052
ATI Radeon HD 2900 XTX, Doomed from the Start
AMD's flagship ATI Radeon HD 2900 XTX fails to usurp the GeForce 8800 GTX's performance crown
The less than stellar performance benchmarks are no surprise to board partners and AMD insiders. Two independent ATI board builders told DailyTech that Radeon HD 2900 XTX will not be a part of their initial portfolios. Given the additional memory cost and bulky card footprint, it seems unlikely the XTX will ever see the light of day.
8800GTX easily frags the HD2900XTX. Looks like Nvidia doesn't even need a 8800Ultra to frag AMD.
No wonder why AMD kept on delaying R600. It's total garbage, just like their processors.
Intel fragging AMD on the CPU front and Nvidia fragging AMD on the GPU front.
AMD is sucking red and losing marketshare. Morgan Stanley won't put up with their crap any longer.
AMD BK 2Q'08
Any R600 based card is not enough to pull AMD's ass out of the financial fire they're in. They're putting far too much focus on this and not enough on getting K10 out the door and into the hands of real IT journalists (as opposed to the fraud that runs this blog)...
Intel Poor Multicore Scaling
Get 3 cores for the price of 4 (lose 1 of 4 cores due to poor scaling).
http://blogs.sun.com/bmseer/tags/scaling
Post a Comment
<< Home