Conroe vs. AMD FX-62

MrWizard6600 said:
right... well this artical says a little differantly. its incomplete, and in the most cpu loaded area (per centage wise) resolution, it pulls ahead, but then 1200X1000 says differantly but i spose thats just cuz of the Nforce 5 chipset.

no doubt conroe will be sweet, but an extra 2,3,4,5 fps isnt that massive an increase, not for me. and i still think that amd's architecture has a little more to offer. but then i like the 2X2mb cache intels got.

You are so ignorant that it hurts.

Firingsquad tested a $316 2.4GHz Conroe vs. FX62 , not a 2.66 one... :D and it proved equal or better than the $1100 FX62.
 
savantu said:
You are so ignorant that it hurts.

Firingsquad tested a $316 2.4GHz Conroe vs. FX62 , not a 2.66 one... :D and it proved equal or better than the $1100 FX62.


suggest retail prices mean nothing untill u actually see them for sale at said prices.

I would think the old days of new videocards would have taught u that by now.
 
MrWizard6600 said:
no doubt conroe will be sweet, but an extra 2,3,4,5 fps isnt that massive an increase, not for me. and i still think that amd's architecture has a little more to offer. but then i like the 2X2mb cache intels got.
The E6600 is a $300 CPU. The FX-62, like all FX processors, will be $1500 at launch. 1-2% peformance difference, 80% price difference. I think you get the idea ;)

dajet24 said:
suggest retail prices mean nothing untill u actually see them for sale at said prices.
Intel never sells for full MSRP, look at the Pentium D 930 if you want proof.
 
savantu said:
You are so ignorant that it hurts.

Firingsquad tested a $316 2.4GHz Conroe vs. FX62 , not a 2.66 one... :D and it proved equal or better than the $1100 FX62.

A good Opteron 165 can come close to FX62 speeds on air, for around $300. I think everyone agrees that the FX line has NEVER been a "bang for the buck" CPU. Nor has the Intel EE line.

That said, I want to see more independant game benchmarks, synthetics don't mean much...the high resolution Quake results shown on Firing Squad seem to prove that.

I could give a hoot how fast a Conroe is at 800x600, or how many milliseconds faster it encodes an MP3, I want to know if it's any real help for a high resolution gaming environment.
 
Croak said:
A good Opteron 165 can come close to FX62 speeds on air, for around $300. I think everyone agrees that the FX line has NEVER been a "bang for the buck" CPU. Nor has the Intel EE line.
Not allowed to compare stock to overclock. Crank that FX62 past 3GHz, THEN compare it to your Opteron.

Croak said:
That said, I want to see more independant game benchmarks, synthetics don't mean much...the high resolution Quake results shown on Firing Squad seem to prove that.
Although I too would like to see more independant sources (come on [H]!), you must remember that FiringSquad used a lower-end (read: cheaper) Conroe than the other benchmarkers did.

Croak said:
I could give a hoot how fast a Conroe is at 800x600, or how many milliseconds faster it encodes an MP3, I want to know if it's any real help for a high resolution gaming environment.
Gotta argee with you on the resolution thing, who the hell plays at 800x600 with a CPU like that? Unfortunately, the only thing thats going to speed up higher resolutions is the graphics card.
 
Croak said:
A good Opteron 165 can come close to FX62 speeds on air, for around $300. I think everyone agrees that the FX line has NEVER been a "bang for the buck" CPU. Nor has the Intel EE line.

That said, I want to see more independant game benchmarks, synthetics don't mean much...the high resolution Quake results shown on Firing Squad seem to prove that.

I could give a hoot how fast a Conroe is at 800x600, or how many milliseconds faster it encodes an MP3, I want to know if it's any real help for a high resolution gaming environment.

$325 - opteron 165, overclocked or stock gets its ass handed to it=P
There are benchmarks all over the web, independant and whatever. Guys at Xtremesystems.org have them and are overclocking and running them in the ground.

The 800 X 600 benchmarks remove the maxed out video card out of the equation. Meaning that if there were better Video cards, you'd get better performance from the more powerful CPU.

All Firingsuad did was Max out one VIDEO card. The others show what Crossfire (Harder to max out) can do. Yes FS's tests were Valid but also NOT meant as an across the board flat result. Many say id has been pro AMD since Doom 3. That's why it might be wise to test at least 4 diffrent Game engines. All Game optimizations aren't the same and always favor different HW. Farcry favors Intel.

http://www.hexus.net/content/item.php?item=5692&page=3

Independent and the Conroes are handicapped LOL! This has both E6600 and E6700 BTW. Conroe is clock for clock faster but we should get a 165 Opteron that's at 1.8GHz and Half the cache? You say mem controller, I say VT and smart memory access and so on. Now let's see stock vs stock LOL!

Overclocking, FCG, Victor Wang, Team Japan, Coolater, Fugger and others are independent all have gotten over 3GHz overclocks. After dud 3000+ and 3500+ Venices cores, I don't trust AMD fans enough to pay $330 for something that will end up slower than stock after it is overclocked to 2.8GHz.

$659 - opteron 180 is the same speed at the default clock of 2.4GHz. not even close.
$1124 - opteron 185 is just totally LAME LOL!

After AMD has to cut prices to more realistic prices or as the flood of folks dumpping their Optys hit the market, I might upgrade my 3500+ with one.
 
Croak said:
A good Opteron 165 can come close to FX62 speeds on air, for around $300. I think everyone agrees that the FX line has NEVER been a "bang for the buck" CPU. Nor has the Intel EE line.

It gets spanked if it were at FX-64 speeds.

Croak said:
That said, I want to see more independant game benchmarks, synthetics don't mean much...the high resolution Quake results shown on Firing Squad seem to prove that.

Hi res maxes out the video card and limits all systems. Nothing to with Processor performance. Yes, all cards don't bog down trying render the Dark assed id games. I actually like Quake 4 Hehehehe.

Croak said:
I could give a hoot how fast a Conroe is at 800x600, or how many milliseconds faster it encodes an MP3, I want to know if it's any real help for a high resolution gaming environment.

Then you don't know much about benchmarketing do ya'? In this case, 800 X 600 shows what higher settings would do with Crossfire or SLI. Id + ATI + Intel is not that good of an Optimization match. Just matching AMD running that software is a feat in and of itself. Sort of like when AMD caught Intel running Quake 3.
 
Donnie I like his argument . Because with that argument my wifes 2 year old 3.2c is as good as Conroe or FX-62 at high res. So all this time Intel has been in the game. Its just that now a differant group of people have seen the light.
 
$BangforThe$ said:
Donnie I like his argument . Because with that argument my wifes 2 year old 3.2c is as good as Conroe or FX-62 at high res. So all this time Intel has been in the game. Its just that now a differant group of people have seen the light.
Now that this argument works for AMD. :rolleyes: When AMD had the crown this argument was shoved under the rug. :D
 
AMD still has the crown . When conroe is released than and only than will there be a coronation
 
$BangforThe$ said:
AMD still has the crown . When conroe is released than and only than will there be a coronation

They don't have it a xtremesystems.org/forums... :p

Terra - Conroe is a fact...the rest is semantics...
 
Terra said:
They don't have it a xtremesystems.org/forums... :p

Terra - Conroe is a fact...the rest is semantics...


its kinda funny its only taken Intel what 3 years to retake the gaming crown.

awful proud for goliath finally catching david,, i dont think they will hold it till 2007.
 
Terra said:
They don't have it a xtremesystems.org/forums... :p

Terra - Conroe is a fact...the rest is semantics...
I know right.

Why does everyone have a problem with this? Just because Intel won a round, everyone is all tits up about it...

Conroe is a superior architecture compared to anything else. And Chips are already out there, and hell, we will be able to buy them this month. I dont see the debate here?
 
dajet24 said:
its kinda funny its only taken Intel what 3 years to retake the gaming crown.

awful proud for goliath finally catching david,, i dont think they will hold it till 2007.

What a poor memory you have...
Intel reigned supreme for YEARS before AMD.
Try counting from 1983 and then do the math.
I actually am kinda impressed how FAST they whipped up and substitute for the failed(due to unforseen MHz problems) Netburst.
And by "failed" I mean that it failed to achive +4Ghz...their aim was ~10GHz.

Terra - Intel is massive...don't expect them to stumble to many times ;)
 
USMC2Hard4U said:
I know right.

Why does everyone have a problem with this? Just because Intel won a round, everyone is all tits up about it...

Most people I see "rage" about seem like people that can't remember from PIII and back.
And when looking at Intel you must not forget their chipsets.
Intel is so much more than desktops CPU's
Intel has been here since 1968 and I see no sign of them fading away now ;)

Conroe is a superior architecture compared to anything else. And Chips are already out there, and hell, we will be able to buy them this month. I dont see the debate here?

You forget, all what we have seen are "Intel rigged" benches ;) :p :rolleyes:
*slams head into the wall*

Terra - But I am ready to visit a certain AMD Zealot blogger when Conroe is retail...MUHAHAHAHA!!!.*chough*
 
Terra said:
What a poor memory you have...
Intel reigned supreme for YEARS before AMD.
Try counting from 1983 and then do the math.
I actually am kinda impressed how FAST they whipped up and substitute for the failed(due to unforseen MHz problems) Netburst.
And by "failed" I mean that it failed to achive +4Ghz...their aim was ~10GHz.

Terra - Intel is massive...don't expect them to stumble to many times ;)
So true. People have completely blown AMD out of proportions. AMD was a one-hit-wonder (A64), and their fanbase simply cannot accept this fact. Intel has reigned supreme since 1983, with the exception 2004-2006 (A64's 15 minutes of fame). Intel's back. Get over it.
 
InorganicMatter said:
So true. People have completely blown AMD out of proportions. AMD was a one-hit-wonder (A64), and their fanbase simply cannot accept this fact. Intel has reigned supreme since 1983, with the exception 2004-2006 (A64's 15 minutes of fame). Intel's back. Get over it.

Actually, I don't think you can write off AMD that quickly.....

However, if the processor wars continue and AMD and INTEL keep one-upping each other, it can only be better for us.... competition is a healthy thing and it pushes for better, cheaper, faster technology. I hope AMD counters with something awesome and Intel eventually does the same. Monopolies suck! (I am going to keep my mouth shut about MS here....)
 
kilgore777 you can say microsoft sucks.
I AGREE

And I have stated that the compotition between AMD and INTEL will only do all of us good
with faster, better and cheaper things.

Right now I want a conroe...next year it might be an AMD...but to me AMD's biggest fault is its pricing.

I tried to buy a 170 last november and it went up $150 in 3 weeks while I was trying to decide on mb and what not. Simply because the little co can't make enough of anything.
Wait till they try and sell 65 nm processors....the fan boys won't keep there mouths shut about kick ass....and watch it go up and up in price since they will only be able to make 200 a month. LOL


sparks
 
coldpower27 said:
Now that this argument works for AMD. :rolleyes: When AMD had the crown this argument was shoved under the rug. :D


Not by me it wasn't, everything has a time and place. Put it another way, I want to see point at which each card processor combo maxes out. I want to see this down with different games because each engine will play to different processors weakkness and strenghts. What that person said was take one game, max it out and that's how they all will do, we know that's bogus. FRAPS is better than Time Demos anyway.

One thing the Anandtech quick review showed, if Science mark isn't dismissed by now, they don't get it. As anyone cross checks it against everything else ran, it becomes a Bigger joke than ever. Sorry if you guys are reading this. :( Was Anyone surprised C2D lost that test, honestly?
 
InorganicMatter said:
So true. People have completely blown AMD out of proportions. AMD was a one-hit-wonder (A64), and their fanbase simply cannot accept this fact. Intel has reigned supreme since 1983, with the exception 2004-2006 (A64's 15 minutes of fame). Intel's back. Get over it.

The last two Quarters of 2005 to the first two of 2006. Try AMD/A64 was the king of Games but there were still a lot of areas Intel wasn't hurting in at all. X2 changed everything, not the low yield highly priced Athlon64's. Prescott hurt Intel more than any AMD processor except maybe X2. Many folks in this market also bought Pentium M even after X2 and many more bought Yonah. The problem was many consumers and AMD fans just didn't get it. As I told duby229 Yonah owners saw Conroe coming when others didn't.

Sure webmasters could do like Extremetech and test one AMD friendly, One Intel friendly app, no multitasking, 9 games and call A64 a run away winner. Kyle and the gang is smarter than that. That's what caused of a lot of the Fan Flame Fests. I say that while I bought a 3500+/939 myself. With both at stock, it crushes my 2.6C with ANY game. Yet, it runs MovieMaker 10, Video Impressions, WMP 10, Creative WAVE creator, and a crap load of other multi-media software slower, some, a lot slower than my 2.6C. The 3500+ also refuses to be overclocked an stay stable. My 2.6C can do 3.52GHz.
 
Donnie27 said:
As I told duby229 Yonah owners saw Conroe coming when others didn't.
Yonah adopters made out like bandits. Merom is 100% compatible with the Yonah platform, all it needs is a BIOS flash...lucky dogs.
 
Donnie27 said:
...I want to see this down with different games because each engine will play to different processors weakkness and strenghts. What that person said was take one game, max it out and that's how they all will do, we know that's bogus. FRAPS is better than Time Demos anyway.

If you're referring to my post, please read it again...we seem to agree. When I said "high resolution gaming environment", I meant a wide gamut of games, not just Q4...just that Q4's high resolution results seemed to support a theory that synthetics were not telling the whole story.

I'm not anti-Conroe in the slightest, I just want to see real world gaming results on retail Conroe chips and retail motherboards, compared to similar AMD systems, without all the hype and drama.

Off topic a little bit, (and I'm sure it's been said before), but the Intel, AMD, ATi and Nvidia subforums here are getting an awful lot like Republicans vs Democrats, which is a shame. I don't remember it being this bad a few years ago when I was reading the Hard Forums regularly (just started visiting again a few months ago).
 
Donnie27 said:
The last two Quarters of 2005 to the first two of 2006. Try AMD/A64 was the king of Games but there were still a lot of areas Intel wasn't hurting in at all. X2 changed everything, not the low yield highly priced Athlon64's. Prescott hurt Intel more than any AMD processor except maybe X2. Many folks in this market also bought Pentium M even after X2 and many more bought Yonah. The problem was many consumers and AMD fans just didn't get it. As I told duby229 Yonah owners saw Conroe coming when others didn't.

Sure webmasters could do like Extremetech and test one AMD friendly, One Intel friendly app, no multitasking, 9 games and call A64 a run away winner. Kyle and the gang is smarter than that. That's what caused of a lot of the Fan Flame Fests. I say that while I bought a 3500+/939 myself. With both at stock, it crushes my 2.6C with ANY game. Yet, it runs MovieMaker 10, Video Impressions, WMP 10, Creative WAVE creator, and a crap load of other multi-media software slower, some, a lot slower than my 2.6C. The 3500+ also refuses to be overclocked an stay stable. My 2.6C can do 3.52GHz.

Well, YMMV. Intel has a strong hold on video and will likely maintain a lead for that. Then again, a lot of these programs are optimized for computing on an intel cpu (and well, Intel makes the best compiler for c++ period.) Ultimately that sort of thing is dependent on what you find more important. It doesn't matter to a gamer if the intel processor encodes video faster than the a64. It doesn't matter to the video editor that the a64 has better performance in games than the Intel.

That said, how do those two compare in those matters? A lot of people say 'a lot slower' or 'marked improvement' when the difference in the real world is quite negligble.

Anyway, people keep saying that competition is good for us but really, it's hard to call this a real 'competition.' If you keep up to date with the news, Japan found that they are using ilelgal tactics to maintain their monopoly, Europe and South Korea are also investigating this. Same thing is happening in the U.S.

Not to say that is definitely a reason, but maybe Intel can lower their prices because they are abusing their monopoly?

...Not that it really matters, seeing as Conroe is gonna be a hit, regardless.
 
Least said:
Well, YMMV. Intel has a strong hold on video and will likely maintain a lead for that. Then again, a lot of these programs are optimized for computing on an intel cpu (and well, Intel makes the best compiler for c++ period.) Ultimately that sort of thing is dependent on what you find more important. It doesn't matter to a gamer if the intel processor encodes video faster than the a64. It doesn't matter to the video editor that the a64 has better performance in games than the Intel.

No, Hyperthreading may not have acted like two processors but it did have at least the affect of 1.5 Processors. Many of the apps lend themselves to multi-threads or to have gaps in the data filled and more effectively processed. These same apps run faster on an X2 for example.

Least said:
That said, how do those two compare in those matters? A lot of people say 'a lot slower' or 'marked improvement' when the difference in the real world is quite negligble.

I call doing a very large file in 20 mins instead of 37 mins a large difference. 80 FPS to 53 is an ass kicking=P That’s not negligible.

Least said:
Anyway, people keep saying that competition is good for us but really, it's hard to call this a real 'competition.' If you keep up to date with the news, Japan found that they are using ilelgal tactics to maintain their monopoly, Europe and South Korea are also investigating this. Same thing is happening in the U.S.

Not to say that is definitely a reason, but maybe Intel can lower their prices because they are abusing their monopoly?

...Not that it really matters, seeing as Conroe is gonna be a hit, regardless.

Intel Japan is not Intel Europe or Intel whatever. But there is no way in hell I'd add Europe to the point you're trying to make, that's about 180 degrees wrong. It is Intel who should be complaining about Europe.

The Germans Built FAB -30 and a lot of the loan was written off, never meant to be paid back. The Germans and the EU updated FAB. The EU and the Germans then Spent 2.6 Billion to build FAB 36, will again spend another billion to expand it. Then they used another 1.5 Billion to upgrade FAB 30 to FAB 38. The turn around and Jobs creative is what Driving New York State to try and pull off the same thing. If AMD and Merges with ATI again, the EU and Germans will finance.

Now, since when in the Hell should any American have to compete with the Financial Wing of the European Union and the German Government as well as the German State of Saxony? Look up AMD Saxonia GMBH?
 
Croak said:
If you're referring to my post, please read it again...we seem to agree. When I said "high resolution gaming environment", I meant a wide gamut of games, not just Q4...just that Q4's high resolution results seemed to support a theory that synthetics were not telling the whole story.

I'm not anti-Conroe in the slightest, I just want to see real world gaming results on retail Conroe chips and retail motherboards, compared to similar AMD systems, without all the hype and drama.

Off topic a little bit, (and I'm sure it's been said before), but the Intel, AMD, ATi and Nvidia subforums here are getting an awful lot like Republicans vs Democrats, which is a shame. I don't remember it being this bad a few years ago when I was reading the Hard Forums regularly (just started visiting again a few months ago).

Originally Posted by Croak
That said, I want to see more independant game benchmarks, synthetics don't mean much...the high resolution Quake results shown on Firing Squad seem to prove that.

Quake results are only valid for Quake=P It only proves anything for that ONE Game, not even other Quake engine Games. I said run at least 4 different Games/Game engines. Firing Squad also used a 2.4GHz Conroe and knowing them, it was more than likely not set up worth a fart. Quake has no relevance to Unreal or Battlefield 2, or Painkiller, MoH or CoD2, that's what I meant.

Like many others, they painted themselves into a corner by saying "We don't see Intel catching up with AMD until they move to an IMC". Rick and Nathan Geek, Joel at Sudhian, Wes Finak at Anand, most of Tom's folks, the guys at Ace's and many many other Webmasters said the same thing. Note, I'm NOT meantioning AMD Fans sites who said the same.

I'm not calling you Anti Intel or an AMD Fan or etc........ I'm just saying that's NOT how you read benchmarks or use them to determine speed. I still learn quite a bit from many here, even those I disagree with.

Choking the system proves almost nothing in that it devalues an A64 FX-57 vs a 4000+ if they both have one 7900GT max out from His res settings. Many would say 1280 X 1024 is their max gaming setting 1600 X 1220 with AA and AF turned up? First of all, you don't need AA when using 1600 X 1200. Maybe 2X AF. Then using SLI or Crossfire change the results again.
 
Croak said:
A good Opteron 165 can come close to FX62 speeds on air, for around $300. I think everyone agrees that the FX line has NEVER been a "bang for the buck" CPU. Nor has the Intel EE line.

That said, I want to see more independant game benchmarks, synthetics don't mean much...the high resolution Quake results shown on Firing Squad seem to prove that.

I could give a hoot how fast a Conroe is at 800x600, or how many milliseconds faster it encodes an MP3, I want to know if it's any real help for a high resolution gaming environment.
If your going to create a GPU limited scneario you can apply this same issue to Pentium D vs Athlon 64x2, the Pentium D will still perform adequately because the bottleneck is the GPU and not the processor itself.

Not to mention that FiringSquad testing the E6600 which will retail for 316US, and don't go giving me that they will gouge 200% on this processor to make it comparable in price to the FX. :rolleyes: As well if you wish to pick a processor and overclock it and use the assumed overclock as a comparison you can also do that with a processor form the Intel cam pas well.

If the stock E6600 can match the stock Athlon FX 62, that's already good enough for me.
 
Donnie27 said:
Not by me it wasn't, everything has a time and place. Put it another way, I want to see point at which each card processor combo maxes out. I want to see this down with different games because each engine will play to different processors weakkness and strenghts. What that person said was take one game, max it out and that's how they all will do, we know that's bogus. FRAPS is better than Time Demos anyway.

One thing the Anandtech quick review showed, if Science mark isn't dismissed by now, they don't get it. As anyone cross checks it against everything else ran, it becomes a Bigger joke than ever. Sorry if you guys are reading this. :( Was Anyone surprised C2D lost that test, honestly?
Sciencemark 2.0 is about the only benchmark in existance that I am aware of where AMD has a chance of outperforming Core Architecture period.

It's because the K7/K8 design has a total of 3 floating point units while Core 2 Duo I believe only has 2.

With any modern game we can expect the usual results at 12x10/16x12 with 4XAA/16XAF, yes AA and AF are required still. Some people can't satnd jaggies...
You will have the GPU limiting the processors and both systems performing identically.

Once you allieviate this bottleneck with SLI/Crossfire you will shift the limitation back to the CPU's and Core 2 Duo will once again coem out on top. Though you cna always run 20x15/25x16 and have the GPU limitation kick in again. :rolleyes:

When the bottleneck is the GPU then you can use a Pentium D and you will still get the same results nevermind the Core 2 Duo vs Athlno 64x2.
 
i think its the minimum fps is where conroe will really shine. sure they will boost maximum fps with crossfire/sli but for us sane foke ( :D ) i think minimum fps is going to see a massive boost with conroe over amd 64.

on a side note ive started purchasing the hardware for my conroe rig.

first up was the ram. g.skill 2GB ddr2-800 4-4-4-12. Not bad for 200 bucks. DDR2 prices have come down quite a bit.

http://www.newegg.com/Product/Product.asp?Item=N82E16820231073
 
coldpower27 said:
Sciencemark 2.0 is about the only benchmark in existance that I am aware of where AMD has a chance of outperforming Core Architecture period.

It's because the K7/K8 design has a total of 3 floating point units while Core 2 Duo I believe only has 2.

With any modern game we can expect the usual results at 12x10/16x12 with 4XAA/16XAF, yes AA and AF are required still. Some people can't satnd jaggies...
You will have the GPU limiting the processors and both systems performing identically.

Once you allieviate this bottleneck with SLI/Crossfire you will shift the limitation back to the CPU's and Core 2 Duo will once again coem out on top. Though you cna always run 20x15/25x16 and have the GPU limitation kick in again. :rolleyes:

When the bottleneck is the GPU then you can use a Pentium D and you will still get the same results nevermind the Core 2 Duo vs Athlno 64x2.

Not that K8 out performed C2D but by how much. AMD didn't hire one of the writers out of fairness, just as Intel pitched in for Incoming Forces that was buildt from the ground up with SSE in mind and used whenever possible. Software can be written to perform better on either processor=P Even Games. Incoming Forces Ran Faster on a Willy Clock for clock than any Athlon of those days. The difference is, this is a benchmark is a test to give the rewiewer and viewer an Idea of performance on similar apps. That's why some folks say 3Dmarks suck because it doesn't use real world Game engines. Is Sciencemark reflecting true realworld Science software?

I've played with crossfire and know a little AF can be used but AA is a waste on the already sweet Graphics I saw. Now go and try to play Halflife 2 online at 1600 X 1200 with AF and AA or Battlefield 2, MoH, Painkiller or etc.., please try it? I have Cable and about 400K. ET-Quake Wars will be only onLine and it is spectacular. See the MaxPC nVidia insert-ad for Details?

I'm all for doing more tests at a more realistic 1280 X 1024 with a little AF. That's the setting I see more than anything. Or 1024 X 768 with AF and AA where the Eye-Candy difference really matters.
 
Donnie27 said:
Not that K8 out performed C2D but by how much. AMD didn't hire one of the writers out of fairness, just as Intel pitched in for Incoming Forces that was buildt from the ground up with SSE in mind and used whenever possible. Software can be written to perform better on either processor=P Even Games. Incoming Forces Ran Faster on a Willy Clock for clock than any Athlon of those days. The difference is, this is a benchmark is a test to give the rewiewer and viewer an Idea of performance on similar apps. That's why some folks say 3Dmarks suck because it doesn't use real world Game engines. Is Sciencemark reflecting true realworld Science software?

I've played with crossfire and know a little AF can be used but AA is a waste on the already sweet Graphics I saw. Now go and try to play Halflife 2 online at 1600 X 1200 with AF and AA or Battlefield 2, MoH, Painkiller or etc.., please try it? I have Cable and about 400K. ET-Quake Wars will be only onLine and it is spectacular. See the MaxPC nVidia insert-ad for Details?

I'm all for doing more tests at a more realistic 1280 X 1024 with a little AF. That's the setting I see more than anything. Or 1024 X 768 with AF and AA where the Eye-Candy difference really matters.
Well of course since Sciencemark uses Floating Point alot more then usually, the K8 architecture will pretty much shine in it, from what I recall Quake 3 ran quite well on Willamette Pentium 4 even for some wierd reason. There will always be applications suited to certain architectures.

I personally go by actual applications I use as a gaude for performance, rather then these synthetics benchmarks.

Just because you don't feel like you need AA at 16x12 doesn't mean you can just go blanket wide statement that nobody does or what is best for everybody, the industry standards for current high end video hardware are 10x7 to 16x12 with 4XAA/8 or 16AF for Single GPU High end benchmarks as well as 12x10 with 4XAA/8 or 16XAF and upwards for SLI and Crossfire.

Typically benchmarks benchmark the resolutions with no AA and AF and the 4XAA/8 or 16AF setting so you can extrapolate pretty easily in what range your selected hardware will perform at that particula resolution.

There a reason there isn't a comparison above 4XAA for the most part not because there aren't modes above that, there are but the implementation used are different so aren't directly comparable anymore.

Not exactly fair asking me to look for evidence I have to jump through hoops to get, I don't want to buy a MaxPC Magazine nor have to run down to the store to see if they even carry it.

You shouldn't use the words 12x10 as realistic, just a setting that is relevant to you, and that you play personally. Same thing with 10x7 comment.
 
coldpower27 said:
Well of course since Sciencemark uses Floating Point alot more then usually, the K8 architecture will pretty much shine in it, from what I recall Quake 3 ran quite well on Willamette Pentium 4 even for some wierd reason. There will always be applications suited to certain architectures.

I personally go by actual applications I use as a gaude for performance, rather then these synthetics benchmarks.

Just because you don't feel like you need AA at 16x12 doesn't mean you can just go blanket wide statement that nobody does or what is best for everybody, the industry standards for current high end video hardware are 10x7 to 16x12 with 4XAA/8 or 16AF for Single GPU High end benchmarks as well as 12x10 with 4XAA/8 or 16XAF and upwards for SLI and Crossfire.

Typically benchmarks benchmark the resolutions with no AA and AF and the 4XAA/8 or 16AF setting so you can extrapolate pretty easily in what range your selected hardware will perform at that particula resolution.

There a reason there isn't a comparison above 4XAA for the most part not because there aren't modes above that, there are but the implementation used are different so aren't directly comparable anymore.

Not exactly fair asking me to look for evidence I have to jump through hoops to get, I don't want to buy a MaxPC Magazine nor have to run down to the store to see if they even carry it.

You shouldn't use the words 12x10 as realistic, just a setting that is relevant to you, and that you play personally. Same thing with 10x7 comment.

Last things first, the 12 X 10 came from a Polls here, at GamePC and even Xtrmemesytems when asked what Settings do you use? It's NOT what I feel, but what polls have consistantly said. No, it is not me trying to speak for everybody but say what those polls showed.

It's not just Floating Point, it's if that Floating Point is RAW or has any SIMD support. AMD does better with RAW, Intel does better with optimized code in most cases. Quake in those days loved Bandwidth and low latency L1 P4 had both. Ace's Hardware had a pretty good debate that pointed out what sucked about SM. Some of Sciencemark makes great use of Bandwdith but even then it set for access, NOT sustained, this again favors AMD.

I already listed some of the apps that do better on either system and I have at least 3 computer that completely different and not counting the wife PC that different as well.

I didn't plan to have you or anyone jump through Hoops.

http://www.splashdamage.com/images/image_04.jpg
http://www.splashdamage.com/images/image_03.jpg
http://www.splashdamage.com/images/image_02.jpg
http://www.splashdamage.com/images/image_01.jpg

http://www.enemyterritory.com/
http://www.enemyterritory.com/main.html
 
Donnie27 said:
Last things first, the 12 X 10 came from a Polls here, at GamePC and even Xtrmemesytems when asked what Settings do you use? It's NOT what I feel, but what polls have consistantly said. No, it is not me trying to speak for everybody but say what those polls showed.

It's not just Floating Point, it's if that Floating Point is RAW or has any SIMD support. AMD does better with RAW, Intel does better with optimized code in most cases. Quake in those days loved Bandwidth and low latency L1 P4 had both. Ace's Hardware had a pretty good debate that pointed out what sucked about SM. Some of Sciencemark makes great use of Bandwdith but even then it set for access, NOT sustained, this again favors AMD.

I already listed some of the apps that do better on either system and I have at least 3 computer that completely different and not counting the wife PC that different as well.

I didn't plan to have you or anyone jump through Hoops.

http://www.splashdamage.com/images/image_04.jpg
http://www.splashdamage.com/images/image_03.jpg
http://www.splashdamage.com/images/image_02.jpg
http://www.splashdamage.com/images/image_01.jpg

http://www.enemyterritory.com/
http://www.enemyterritory.com/main.html
Well from the looks of things those pictures do have AA applied. So it is quite necessary. The shot do look quite good though.

I would also expect the 12x10 resolution to be most common due to the prevalency of 17inch and 19inch LCD 5:4 at affordable prices these days. The issue I am dicussing is the use of AA or AF, not particularly the resolution used. If you do have high end graphics hardware, you also have the cash to have a 20inch LCD of either 16x12 or 1680x1050. It doesn't really matter anyway, if a site benches 12x10 4XAA/8 or 16AF and you get 50+ with this setting you will most certainly get higher as you reduce the image quality enhancing features further.

Like I said most sites already bench 12x10 or 12x9 so this isn't a big issue and you can extrapolate.

Evne still we still need to bench 16x12 4XAA/8 or 16xAF nonetheless as you won't detemine a winner at 12x10 with a 7900 GTX or X1900 XTX those cards would run this resolutions flawlessly. There isn't a need to reduce you AA or AF at all from 4/8 or 16.

Yes the K8 architecture doesn't need optimized code to work well, it's the smart decision to do for a company of AMD's size as they can't use that much resources to work with everyone to try to optimize for their architecture.

Sorry if I came off as too harsh. :)
 
Donnie27 said:
The last two Quarters of 2005 to the first two of 2006. Try AMD/A64 was the king of Games but there were still a lot of areas Intel wasn't hurting in at all. X2 changed everything, not the low yield highly priced Athlon64's. Prescott hurt Intel more than any AMD processor except maybe X2. Many folks in this market also bought Pentium M even after X2 and many more bought Yonah. The problem was many consumers and AMD fans just didn't get it. As I told duby229 Yonah owners saw Conroe coming when others didn't.

Sure webmasters could do like Extremetech and test one AMD friendly, One Intel friendly app, no multitasking, 9 games and call A64 a run away winner. Kyle and the gang is smarter than that. That's what caused of a lot of the Fan Flame Fests. I say that while I bought a 3500+/939 myself. With both at stock, it crushes my 2.6C with ANY game. Yet, it runs MovieMaker 10, Video Impressions, WMP 10, Creative WAVE creator, and a crap load of other multi-media software slower, some, a lot slower than my 2.6C. The 3500+ also refuses to be overclocked an stay stable. My 2.6C can do 3.52GHz.


I'm going to have to blame your lack of overclockability on user error. And don't try calling me a fan boy anymore look at sig :p
 
coldpower27 said:
Well from the looks of things those pictures do have AA applied. So it is quite necessary. The shot do look quite good though.

I would also expect the 12x10 resolution to be most common due to the prevalency of 17inch and 19inch LCD 5:4 at affordable prices these days. The issue I am dicussing is the use of AA or AF, not particularly the resolution used. If you do have high end graphics hardware, you also have the cash to have a 20inch LCD of either 16x12 or 1680x1050. It doesn't really matter anyway, if a site benches 12x10 4XAA/8 or 16AF and you get 50+ with this setting you will most certainly get higher as you reduce the image quality enhancing features further.

Like I said most sites already bench 12x10 or 12x9 so this isn't a big issue and you can extrapolate.

Evne still we still need to bench 16x12 4XAA/8 or 16xAF nonetheless as you won't detemine a winner at 12x10 with a 7900 GTX or X1900 XTX those cards would run this resolutions flawlessly. There isn't a need to reduce you AA or AF at all from 4/8 or 16.

Yes the K8 architecture doesn't need optimized code to work well, it's the smart decision to do for a company of AMD's size as they can't use that much resources to work with everyone to try to optimize for their architecture.

Sorry if I came off as too harsh. :)

OK!
 
I'm going to have to agree with donnie. 1280x1024 pretty much is the standard. I would say I have a relatively high end card (1800xt 512MB with PE bios) and I game at 1280x1024, my monitors native res.
 
sabrewolf732 said:
I'm going to have to blame your lack of overclockability on user error. And don't try calling me a fan boy anymore look at sig :p

I'm going to hve to blame you're comments on BS! :p I thought maybe it was something I was doing too. That's why a I let an AMDroid buddy of mine try it. He has one of those Golden sample Opty 165s at 3GHz and he couldn't overclock my 3500+ either :( Unless you call 180MHz an overclock.
 
sabrewolf732 said:
I'm going to have to agree with donnie. 1280x1024 pretty much is the standard. I would say I have a relatively high end card (1800xt 512MB with PE bios) and I game at 1280x1024, my monitors native res.

Thanks and I can't remember the last time I called you anything negative.
 
Donnie27 said:
I'm going to hve to blame you're comments on BS! :p I thought maybe it was something I was doing too. That's why a I let an AMDroid buddy of mine try it. He has one of those Golden sample Opty 165s at 3GHz and he couldn't overclock my 3500+ either :( Unless you call 180MHz an overclock.

:'( = Donnie :p I must say though, Intel's seem to be easier to overclock and get stable.
 
sabrewolf732 said:
:'( = Donnie :p I must say though, Intel's seem to be easier to overclock and get stable.

Hehehe! I hope you know the BS comment was a joke? Wow it still seems wild seeing you on the "Dark Side" :) Nice overclock you got there.
 
Back
Top