Sign in with
Sign up | Sign in
Your question

Intel's claims Penryn 40% faster in games!

Last response: in CPUs
Share
April 17, 2007 1:11:28 AM

http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.

More benchmark data:

http://techgage.com/viewimg/?img=/articles/intel/beijin...

Penryn is going to be a monster!
April 17, 2007 1:29:02 AM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.


We'll have to wait until we actually see Agena FX. If the perf holds up to the specs I'd say a similarly clocked Agena will be at 80% faster than K8 in games due to the higher RAM speed (supports 1066), faster HT3 bus and dual loads/stores.

I personally will be glad to see an actual benchmark, though I don't doubt AMDs "word." Penryn will be formidable but it seems to me like Intel maybe making too many intros too close together.

Everyone also thought the P4s high clocks would do in K8.
April 17, 2007 1:38:05 AM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.


We'll have to wait until we actually see Agena FX. If the perf holds up to the specs I'd say a similarly clocked Agena will be at 80% faster than K8 in games due to the higher RAM speed (supports 1066), faster HT3 bus and dual loads/stores.
Benches...Links... :p 
Quote:

I personally will be glad to see an actual benchmark, though I don't doubt AMDs "word."

Ain't that the TRUTH!!!
Quote:
Penryn will be formidable but it seems to me like Intel maybe making too many intros too close together.

Which is following their plan as expected... :roll:
Quote:
Everyone also thought the P4s high clocks would do in K8.

They did reach right up there, eventually matching and even besting quite a few once stock speeds got up to 3.6 and 3.8GHz... and then you could still overclock them further...
Related resources
April 17, 2007 1:41:49 AM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.


Well, you don't know how Barcelona performs. You might be just as surprised when they release their K10 benchies on April 23rd (according to a post somewhere on here).
April 17, 2007 1:45:52 AM

But in 6 months, how many revisions and releases will intel have out improving thier performance that these unseen benchmark claims are beating current production C2D's?
So if AMD is indeed 80% faster, will intel be 120% faster?
April 17, 2007 1:57:50 AM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.


Well, you don't know how Barcelona performs. You might be just as surprised when they release their K10 benchies on April 23rd (according to a post somewhere on here).

Here's the issue with this - "they" are going to be AMD releasing benchmarks - not third party - AMD. They should have had ES's in peoples hands by now - they should be doing far more than they are. They're not. So even AFTER the 23rd until Toms, Anand, [H], whoever as long as it's NOT AMD, gets their hands on these parts and releases independent benchmarks then I'll take it all with a with a large grain of salt.

And before any of you AMD fans spout nonsense I feel exactly the same regarding Intel's statements in terms of their upcoming parts....
April 17, 2007 1:59:03 AM

Quote:
But in 6 months, how many revisions and releases will intel have out improving thier performance that these unseen benchmark claims are beating current production C2D's?
So if AMD is indeed 80% faster, will intel be 120% faster?


Well, if Penryn is released before K10 (which is very likely), then AMD will have the advantage of examining a processor so they can base theirs to have higher performance. Oh well, we can't really say anything until we see some benchies! (Wonder how long we've been saying that).
April 17, 2007 2:09:17 AM

Quote:
But in 6 months, how many revisions and releases will intel have out improving thier performance that these unseen benchmark claims are beating current production C2D's?
So if AMD is indeed 80% faster, will intel be 120% faster?


Well, if Penryn is released before K10 (which is very likely), then AMD will have the advantage of examining a processor so they can base theirs to have higher performance. Oh well, we can't really say anything until we see some benchies! (Wonder how long we've been saying that).

I hate when people make statements like this. For crying out loud....

This is quite a hefty claim for Intel to make. Perhaps they mean running Doom 3 at the lowest possible settings on integrated graphics. Who knows.
April 17, 2007 2:15:09 AM

Quote:
But in 6 months, how many revisions and releases will intel have out improving thier performance that these unseen benchmark claims are beating current production C2D's?
So if AMD is indeed 80% faster, will intel be 120% faster?


Well, if Penryn is released before K10 (which is very likely), then AMD will have the advantage of examining a processor so they can base theirs to have higher performance. Oh well, we can't really say anything until we see some benchies! (Wonder how long we've been saying that).

I hate when people make statements like this. For crying out loud....

This is quite a hefty claim for Intel to make. Perhaps they mean running Doom 3 at the lowest possible settings on integrated graphics. Who knows.

Exactly! We can't really say anything until we see some benchmarks!
April 17, 2007 2:39:15 AM

Quote:
Here's the issue with this - "they" are going to be AMD releasing benchmarks - not third party - AMD. They should have had ES's in peoples hands by now - they should be doing far more than they are. They're not. So even AFTER the 23rd until Toms, Anand, [H], whoever as long as it's NOT AMD, gets their hands on these parts and releases independent benchmarks then I'll take it all with a with a large grain of salt.

And before any of you AMD fans spout nonsense I feel exactly the same regarding Intel's statements in terms of their upcoming parts....


Lets keep in mind that AMD is releasing a Server chip. Of course there won't really be any previews of it other than AMD sabre rattling. This is normal for server chips because bussinesses don't run out and buy new server chips because of some review 2 months before the release.

Now as far as the K10 desktop processor benchmarks. There is no release date for K10 desktop chips yet. Hell, if K10 Server chips are in high demand and the desktop market looks bleak for K10 due to price/performance/competition, a K10 desktop processor could still be being tweaked before it will be released. So why would AMD benchmark something they haven't even finalized for market yet???
April 17, 2007 2:39:45 AM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.


We'll have to wait until we actually see Agena FX. If the perf holds up to the specs I'd say a similarly clocked Agena will be at 80% faster than K8 in games due to the higher RAM speed (supports 1066), faster HT3 bus and dual loads/stores.

I personally will be glad to see an actual benchmark, though I don't doubt AMDs "word." Penryn will be formidable but it seems to me like Intel maybe making too many intros too close together.

Everyone also thought the P4s high clocks would do in K8.
Sadly not, AMD has stated that K10 has 80% higher FPU performance (roughly 25% faster in games) than K8-Opt. per core. This is good enough to best the C2D but AMD has skipped a generation so they need to set their sights higher; second place isn't going to cut it. And not everyone is going to want to run 8 core AgenaFX to keep up with the 4 core Penryn.

There is one thing I don't like about, "Penryn is 40% faster .. in games!". What was the resolution 640x480? Was there any eyecandy turned on? Is this determined my measuring useless MAX FPS or by measuring useful MIN FPS (the chunk test)?
April 17, 2007 2:40:17 AM

Quote:
But in 6 months, how many revisions and releases will intel have out improving thier performance that these unseen benchmark claims are beating current production C2D's?
So if AMD is indeed 80% faster, will intel be 120% faster?


Well, if Penryn is released before K10 (which is very likely), then AMD will have the advantage of examining a processor so they can base theirs to have higher performance. Oh well, we can't really say anything until we see some benchies! (Wonder how long we've been saying that).

I hate when people make statements like this. For crying out loud....

This is quite a hefty claim for Intel to make. Perhaps they mean running Doom 3 at the lowest possible settings on integrated graphics. Who knows.

To bench a CPU on running game code, this is the only way to do it ... otherwise you are not benching the CPU you are benching the GPU.

This is true.

I'm not much of a PC gamer (360 all the way), I'm more interested in video / audio encoding power and other tasks. So, in my mind, getting 120fps instead of 110 seems like a bit of a moot point to me. I don't mean to sound like Baron, though.
April 17, 2007 3:10:39 AM

no more talk clamining % increase we want the benchmarks....................... lets see who got higher performance
April 17, 2007 3:14:19 AM

Perhaps more important from an AMD point of view is that Intel will FINALLY (my god it is f&*king time) get rid of the Netburst based chips for 4 sockets and up.

The 8xxx series Opterons are AMD's best cash cow, they still go for well over $1000 a pop, because the combination of Hypertransport and the fact that netburst utterly sucks means that in 4s/8 cores and up, Intel is currently getting destroyed.

4 Woodcrest/Clovertowns on a 4 socket/4 FSB board + 64 meg snoop filter, while certainly in no danger whatsoever of winning a beauty prize for technical elegance, will at least be back in the game, perhaps even marginally ahead, performance wise.

~~~~~

Due in Q3 of this year is also Intel’s Caneland platform with Tigerton dual- and quad-core processors. The CPUs will replace the aging Xeon MP 7100 series and will not only complement Intel’s 65 nm Core architecture deployment, but also allow Intel to regain market shares it has lost to AMD in the lucrative MP server segment.

By the end of the year, Intel will also be offering a dual-socket consumer PC platform. Depending on your view, this could either be seen as a reaction to AMD’s dual-socket QuadFX (4x4) platform or as a logical step to offer what some enthusiasts have been doing for some time - to take a Xeon server board and use Woodcrest or Clovertown processors to build a system with two dual- or quad-core processors. Intel calls this project “Skulltrail,” which will support two Core processors with up to eight cores. The platform will also run up to four graphics cards.
April 17, 2007 3:39:57 AM

lol!

40% faster - 40% faster in which game, a which resolution????


wow this just happens to match the statement by amd that their new processors are 40% faster then c2d!

Get it - its bs!
April 17, 2007 3:44:06 AM

damn mayn, I'm here waitin to purchase parts for somethin new and all this ill ass stuff is jus waitin to be introduced to the market...

decisions decisions
April 17, 2007 3:48:39 AM

It would appear that a ~14% faster clock speed penryn (3.33ghz vs 2.93ghz) is 40% faster in whatever game they claim. So this makes it ~26% faster clock for clock than the current intel core 2 quad cores am I correct? Will this be enough to best barcelona from the information we have recieved that is, which looks better so far...I havent had much time to look over the situation! Any thoughts?

Best,

3Ball
April 17, 2007 4:16:57 AM

AMD has claimed Barcelona will be 40% faster on one FP benchmark. Yes, the same one that Henri Richard stood around complaining that was too old to use.

Is everyone else sick of AMD's puerile whining about "open and fair compeition" and how Intel only shows benchmarks that favour their products when AMD does exactly the same thing?! AMD could save some cash by firing Henri Richard for sure!
April 17, 2007 5:18:29 AM

Quote:

There is one thing I don't like about, "Penryn is 40% faster .. in games!". What was the resolution 640x480? Was there any eyecandy turned on? Is this determined my measuring useless MAX FPS or by measuring useful MIN FPS (the chunk test)?


Couldn't agree more.
April 17, 2007 7:53:08 AM

The keypoint being - how long until applications start to take advantage of SSE4?
April 17, 2007 8:06:57 AM



1st, Penryn has about 10% frequency advantage.
2nd, the "211%" percent faster is based on optimized codes which should not be treated as other benches. I will need a SSE-3 optimized encoding application to confirm the SSE code performance.
April 17, 2007 8:35:22 AM

Quote:
Sadly not, AMD has stated that K10 has 80% higher FPU performance (roughly 25% faster in games) than K8-Opt. per core.

FP performance has nothing to do with GAMES performance!
Games CPU performance depends of its AL and its SSE performance(90% of the instructions).
K10 will perform faster in games for sure, but not faster than Core2, clock for clock.
April 17, 2007 8:51:37 AM

and tell me why I need it when I have yet to find a game that can use more than half my 4200x2?
to use that you will need a r600+
April 17, 2007 9:09:48 AM

Quote:
That aside, this is indeed impressive. Aside from the roughly 10% clock improvement, we observe a significant improvement in is a few cases -- but question -- has the 2707 build of HL2 been SSE4 opt? Most certainly DivX has been. The most interesting number, if we take the data as speculative, is the Cinebench data which shows a huge improvement (indicating the FP improvements Intel indicated).
Jack


It would not be surprised that the FPU performance will increase in the coming two or three generations. I am sure that Intel will tackle its FPU issue just as AMD is tackling its SSE issue. :wink:
April 17, 2007 10:56:16 AM

Quote:
lol did you not see the half life 2 score? do you know what divx is?
what size resoultion do you run games at?


I run them @ 1280 x1024 and hl2 is no where near using up my cpu something like 60-70% core 1 and 10 % core 2
April 17, 2007 10:57:06 AM

I could have sworn that x2 performance is more in line with 100% than 211%. Where did that number come from anyway?
April 17, 2007 12:18:37 PM

try supreme commander then you will want a faster CPU


and im not going to comment on the supposed benchmarks

roll on apr 23rd
April 17, 2007 1:07:31 PM

Quote:
...Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.
...Penryn is going to be a monster!


What monster? I could have sworn that I saw 3.6 or 4.0 GHz predicted for Penryn on forumz?? What is the metal gate advantage 26% clock increase?? That won't be enough to kill anything if Barcelona has slightly more IPC.
April 17, 2007 1:34:19 PM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.

More benchmark data:

http://techgage.com/viewimg/?img=/articles/intel/beijin...

Penryn is going to be a monster!

Actually there is a fierce speculation battle between AMD and Intel in full expansion, so I'd not be surprised if this 40% was for a single threaded game , taking advantage of Penryn's automatic core overclocking; you might have the core in question running @ 3.8GHz instead of 3.33 :D 
But then again, this is MY speculation and I'd not fall in this percentage game till we have some real benchmarks; too much has been said and speculated on so we'd better be patient.
April 17, 2007 1:41:57 PM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.

More benchmark data:

http://techgage.com/viewimg/?img=/articles/intel/beijin...

Penryn is going to be a monster!


So....the same fanboys that are knocking AMD's claims of a 40% faster Barcelone, are now pimping Intels claims of a 40% faster Penryn. ROFL.
You gotta love the nerd boy mentality in these forums.
Just like AMD's pres said, Intel is engineering product too fast for the software developers and mobo makers to keep up. Therefor they are creating technology that is useless to the mainstream user. There are hardly any games out for dual core let alone quad core yet Intel is steady at work on 8 and 16 core processors as we speak. 8 cores will be announced early next year if not sooner. How stupid is that?Do you love being hated???.....cause you certainly are good at it. :x
April 17, 2007 1:57:40 PM

Quote:

So....the same fanboys that are knocking AMD's claims of a 40% faster Barcelone, are now pimping Intels claims of a 40% faster Penryn. ROFL.
You gotta love the nerd boy mentality in these forums.
Just like AMD's pres said, Intel is engineering product too fast for the software developers and mobo makers to keep up. Therefor they are creating technology that is useless to the mainstream user. There are hardly any games out for dual core let alone quad core yet Intel is steady at work on 8 and 16 core processors as we speak. 8 cores will be announced early next year if not sooner. How stupid is that?


Do you love being hated???.....cause you certainly are good at it. :x

8O Where did that come from?
April 17, 2007 1:58:08 PM

40% faster... HAHAHAHAHA... :lol:  :lol:  :lol: 

leave it to those Intel execs to have a good sense of humor.

They're making a mockery of AMDs previous (and extremely vague) "40% faster" statement.
April 17, 2007 1:59:23 PM

Quote:
http://www.techreport.com/onearticle.x/12272

Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.

More benchmark data:

http://techgage.com/viewimg/?img=/articles/intel/beijin...

Penryn is going to be a monster!


So....the same fanboys that are knocking AMD's claims of a 40% faster Barcelone, are now pimping Intels claims of a 40% faster Penryn. ROFL.
You gotta love the nerd boy mentality in these forums.
Just like AMD's pres said, Intel is engineering product too fast for the software developers and mobo makers to keep up. Therefor they are creating technology that is useless to the mainstream user. There are hardly any games out for dual core let alone quad core yet Intel is steady at work on 8 and 16 core processors as we speak. 8 cores will be announced early next year if not sooner. How stupid is that?
Didnt they learn anything when they tried to win the gigahertz race and failed miserably?
I suppose you did the same thing when AMD introduced the first dual core when there was virtually no software to take advantage of multicore processors, or when they debuted AM64 while few if any computers use 64 bit OS's and few use them still. Or the introduction of the IMC onto a DT platform seeing as how the gains from an IMC will not be seen in the DT field for a few years. Or introducing a multiprocessor multicore gaming platform to the market when few games take full advantage of single processor multicore solutions. You did, didn't you? I mean, you are the paragon of unbiasedness and an example of what a true forum poster should act like.

Hardware has to leap ahead or else the advances to the software that it runs will stagnate. Of course, since it is Intel that is doing the leaping I can see how you wold see this as a foolish waste of R&D. Yes, maybe Intel should have stuck with the Pentium III and just raised the clock speed a bit. I mean, no one really uses multithreaded applications. No one really needs 64bit acsess. Who uses SSE2 - 4 anyways? Some uppity programmers that obviously don't share your enlighten view on the world of computing.

While we're at it, lets get rid of DDR3 and K10, since we don't need to advance at all. Then we will get rid of Cool 'n Quiet and Speedstep, since what use does a slower processor have to slow down?

Please Lady Bytch the Wise, oh holy and truthful and without bias, enlighten all us lamers and fanbois on the proper way to advance computing? Should we tell the fools at AMD and Intel to stop developing smaller process sizes, seeing as you can only add more features to and improve the state of processors?

Don't for a minute think (well, that really isn't asking much from you anyways) that I or others don't cast extreme doubt and prejudice onto Intel's statement. I treat this the same as any other press statement, as a half truth, filled with caveats and conditions. 40% faster than the QX6800 in what games? Under what conditions? Using what platform and equipment? Define 40% faster. You see, it matters not be it AMD or Intel. Marketing is marketing. True, only when viewed in context.
April 17, 2007 2:00:12 PM

I'm just amazed the general tech/financial press doesn't have this forum bookmarked so they can regular updates.... yawn....

This forum has gone from the sublime to the ridiculous.
April 17, 2007 2:04:22 PM

Quote:
...Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.
...Penryn is going to be a monster!


What monster? I could have sworn that I saw 3.6 or 4.0 GHz predicted for Penryn on forumz?? What is the metal gate advantage 26% clock increase?? That won't be enough to kill anything if Barcelona has slightly more IPC.

3.33 GHz is just their current stepping.... 3.5-3.6 GHz is a likely number for this process technology. I do not doubt. A dual core will reach 3.7-4.0 GHz is my guess.

3.33 GHz quad core over a 2.3 GHz quad core ?? This is a Barcelona stopper. If these numbers are true, this is 40% over a 2.93 GHz core 2, and significantly better than Barcelona claims as AMD only claims 40% over a 2.66 GHz.

Well Intel always exagerates and never understates anything. Even the 40% peformace of conroe over k8 turned out to be 20% - 30% at best in real world applications.


Btw the 3.33GHz Penryn has a 1333 bus over QX6800 with has a 1066 wich means the IPC increase is close very little.
April 17, 2007 2:06:05 PM

Much more diplomatic response than mine. :oops:  :wink:
April 17, 2007 2:08:47 PM

Quote:
...Wow! A 3.33 GHZ Penryn is 40% faster than a 2.93 GHZ QX6800 in games! Barcelona is going to get creamed if this is true.
...Penryn is going to be a monster!


What monster? I could have sworn that I saw 3.6 or 4.0 GHz predicted for Penryn on forumz?? What is the metal gate advantage 26% clock increase?? That won't be enough to kill anything if Barcelona has slightly more IPC.

3.33 GHz is just their current stepping.... 3.5-3.6 GHz is a likely number for this process technology. I do not doubt. A dual core will reach 3.7-4.0 GHz is my guess.

3.33 GHz quad core over a 2.3 GHz quad core ?? This is a Barcelona stopper. If these numbers are true, this is 40% over a 2.93 GHz core 2, and significantly better than Barcelona claims as AMD only claims 40% over a 2.66 GHz.

Well Intel always exagerates and never understates anything. Even the 40% peformace of conroe over k8 turned out to be 20% - 30% at best in real world applications.


Btw the 3.33GHz Penryn has a 1333 bus over QX6800 with has a 1066 wich means the IPC increase is close very little.Yeah but, if the aged, archaic, bottleneck of the system..FSB is full, than what should 1333FSB matter? 8)
April 17, 2007 2:09:37 PM

I though Intel said that Penryn would have a 1600HMz FSB?
April 17, 2007 2:10:35 PM

Quote:
Much more diplomatic response than mine. :oops:  :wink:

I want to see the reply to that. I'll be shocked if I do get a reply to that.
April 17, 2007 2:15:48 PM

Quote:
Much more diplomatic response than mine. :oops:  :wink:

I want to see the reply to that. I'll be shocked if I do get a reply to that.She(it?) almost always does drive-by postings, and rarely posts more than once per thread....probably because it is trying to stir up sh*t, and knows the flamage that would ensue.
April 17, 2007 2:22:47 PM

This article/thread is pointless with some cool pictures or benchmarks!
April 17, 2007 2:25:36 PM

April 17, 2007 2:31:08 PM

But still, the answer I would get... I just want to see what Bytch can say to that.

@Wombat
IMG tags only work when the url is ended with a common image extension. VBulletin attachments make it near impossible to post into BBCode with out some thinking... here, let me fix that for you.

Now I hate to be the voice of reason, but what proof do we have that this isn't a complete and total fake?
April 17, 2007 2:40:48 PM

April 17, 2007 2:40:52 PM

Welp ... looks like K10 and Penryn are going to have to compete on price, not performance.

Winner: Consumer 8)
April 17, 2007 2:48:40 PM

Quote:
Winner: Consumer 8)

Looser: BaronMatrix :lol: 
April 17, 2007 2:51:52 PM

Seriously, how do we know that this isn't a fake?
April 17, 2007 2:57:37 PM

Quote:
Seriously, how do we know that this isn't a fake?


Why bother to fake data that's going to be released in 24 hours?
!