• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Conroe debunked

Status
Not open for further replies.
(cf)Eclipse said:
we'll see who has the last laugh here.. but know one thing: there's nothing that pisses me off more than someone disagreeing, saying that someone else is wrong, THEN having the audacity to say something like "i can't say why you're wrong", it just makes you sound like an egotistical bastard who really doesn't know anything at all.
Click to expand...

Sure he could have been nicer but he is under NDA and it could cost him his Job otherwise. He was trying to tell the guy he works for Intel.
 
Donnie27 said:
Sure he could have been nicer but he is under NDA and it could cost him his Job otherwise. He was trying to tell the guy he works for Intel.
Click to expand...
them simply don't say anything if NDA is a concern... by stating what's not true, you're leaving hints to what is true, which can be interpreted as a violation of the NDA depending on how strict it is. ;)
 
(cf)Eclipse said:
them simply don't say anything if NDA is a concern... by stating what's not true, you're leaving hints to what is true, which can be interpreted as a violation of the NDA depending on how strict it is. ;)
Click to expand...

You're probly right!
 
GilmourD said:
rolleye0009.gif


My point is that you're defending a chip that may or may not exist in the form in which you're defending it. We have no proof. We won't until sometime around September to October. [waiting for you to spin this one]
Click to expand...

Donnie27 said:
K8 might not be Conroe but it is still kick-ass.
Click to expand...

I don't think Fugger, Victor Wang, icookie and the others who has ES are lying!
 
Im still finding it incredible that the 2.13 GHz E6400 is kicking ass in super-pi. If Victor Wang got the 2.4 to be cooled passively, than im sure the 2.13 can be very comfortable without a fan, sounds nice for an SFF. What I see going on with the whole !!!!!! thing, is that super-loyal AMD fans are overly in denial that Intel has developed a prize winning chip. I see a lot of this going back and forth about AMD fans either trying to slip in propoganda or bringing out the tiniest imperfections. Wth? Idc who your a fan of, if you want to be treated equally then you need to treat both your fellow members and your fellow companies equally, period.
 
By the pictures HKEPC provided, it looks like it was overclocked on a stock cooler.
 
Bigjohns97 said:
This thread looks pretty stupid after a conroe @ 3.1 hit 16sec superpi.
Click to expand...

isn't that because superpi fits entirely in the 4 meg cache? So really you're comparing the speed of 1meg cache + system memory to 4 meg cache by itself.

There is a reason why benchmark programs are designed to be bigger than the cache size of existing processors. I suspect that we will be seeing "updated" benchmark programs with 4+ meg execution chains to eliminate that unbalance.

Yes, any program that fits entirely in the 4 meg cache will run very fast, that's great. If calculating Pi is your lifeblood, then conroe is for you. what would be interesting is to cache disable both the conroe and athlon64 and then benchmark 'em.

If the larger cache really does make most of the speed difference, and not the processor itself, then all AMD would have to do to "catch up" is move to a larger cache size.
 
This was a 2mb cache conroe not 4. Also, we can wait for 32m PI test that will knock your socks off and blow your theory to hell and gone.
 
Logan321 said:
isn't that because superpi fits entirely in the 4 meg cache? So really you're comparing the speed of 1meg cache + system memory to 4 meg cache by itself.
Click to expand...
The 1 million digit run allocates 8MB of memory. The 32 million digit run for the Conroe 2.4 tested Victor Wang would require a 3.5-3.6 Opteron to match.

If calculating Pi is your lifeblood, then conroe is for you. what would be interesting is to cache disable both the conroe and athlon64 and then benchmark 'em.
Click to expand...
Especially after you disable the memory controller on the A64.
 
^ L O L. What a laugh ;) So many people find some way to bias the Conroe that its acutally quite funny the kind of arguments I hear.
 
Logan321 said:
isn't that because superpi fits entirely in the 4 meg cache? So really you're comparing the speed of 1meg cache + system memory to 4 meg cache by itself.

There is a reason why benchmark programs are designed to be bigger than the cache size of existing processors. I suspect that we will be seeing "updated" benchmark programs with 4+ meg execution chains to eliminate that unbalance.

Yes, any program that fits entirely in the 4 meg cache will run very fast, that's great. If calculating Pi is your lifeblood, then conroe is for you. what would be interesting is to cache disable both the conroe and athlon64 and then benchmark 'em.

If the larger cache really does make most of the speed difference, and not the processor itself, then all AMD would have to do to "catch up" is move to a larger cache size.
Click to expand...

Guess you havent seen the Core based Celerons with 1meg L2.
 
Emission said:
^ L O L. What a laugh ;) So many people find some way to bias the Conroe that its acutally quite funny the kind of arguments I hear.
Click to expand...

Why do you think I have this thread bookmarked ;) *L*

Terra...
 
Logan321 said:
isn't that because superpi fits entirely in the 4 meg cache? So really you're comparing the speed of 1meg cache + system memory to 4 meg cache by itself.

There is a reason why benchmark programs are designed to be bigger than the cache size of existing processors. I suspect that we will be seeing "updated" benchmark programs with 4+ meg execution chains to eliminate that unbalance.

Yes, any program that fits entirely in the 4 meg cache will run very fast, that's great. If calculating Pi is your lifeblood, then conroe is for you. what would be interesting is to cache disable both the conroe and athlon64 and then benchmark 'em.

If the larger cache really does make most of the speed difference, and not the processor itself, then all AMD would have to do to "catch up" is move to a larger cache size.
Click to expand...

Whatever helps you sleep at night.
 
Emission said:
^ L O L. What a laugh ;) So many people find some way to bias the Conroe that its acutally quite funny the kind of arguments I hear.
Click to expand...
i agree actually. i may be biased to amd, but it's not like i'm completely irrational.... :p
 
Shameless Liar said:
Can I have permission to hijack this thread? It's pretty lame now.
Click to expand...

You should have jumped in right after the OP's first post :D

Terra...
 
duby229 said:
I don't work for anybody. I just have a drive to research technology. It is my hobby. I like Intel. I like AMD. I like IBM and Motorola. And I try to keep up with what they do.

Listen. I'm not going to get into the history of fabrication. But if you go back to about the 250nm node for those above mentioned companies. That is where modern fabrication really began. They were using cmos before that , but there were some really amazing advances that accured on that node. And everything that is being done today is based in some form on the same process.

Do a google search. There is plenty of information available.



I guess we'll have to wait until November when it is actually available.
Click to expand...


if you're so knowledgable why don't you inform us??? why do we need to search if you know all this stuff??

Are you refering to the fact that AMD and intel don't make their CPUS completly out of CMOS tech?? and that only they are both finally making the move towards all CMOS cpus???

as far as I know.. AMD cpus have a hire percentage of CMOS tech in a chip.. but that's about it.. i mean.. what are you getting at?? instead of talking msack with no back.. why don't you refer a few sites.. i mean what exactly are you talking about???
 
Donnie27 said:
QFT!
Click to expand...
yup, even if we agree to disagree, i know you try to see where i'm coming from :cool:

as a side note, this thread is basically dead?
unless we want to talk about the crazy folks over at anandtech who think they got ahold of a 2.66ghz 5000+ :rolleyes:
 
Donnie27 said:
Of course they're the leader. But how many times have you read things written by AMD worshippers saying AMD wouldn't do such a thing?

I've already said i965 and E6600 or bust for me.

Nov-5. I was talking to Sparks and we were planning to get Dual Core Opterons. We watched in shock as the prices jumpped. I bought the 3500+ and he put off buying anything. No, I don't think the 3800+ has ever been worth its price. Since Northwood, the only thing Intel had of any interest to me is Presler and maybe Cedar Mill. Now if they'd launched last summer or Fall, who knows.

For my Conroe system, whenever it ships, I had planned on nVidia's 7900GT but since I want an Intel based chipset and they BLOCK SLI, I think I'll used ATI, again. Since many nVIdia fans are also AMD (SNAP), I don't get the lack of outrage.

All 4 of PC Perspective's Leader board systems are AMD. They're hardly AMD Fans though. Here's they say;

I think he should have said under what the 4800+ costs by about $100 and across the board LOL!
Click to expand...
Well I typically want to ignore the statements of AMD fanatics, as they will justify any actions of AMD as good, while anything Intel does is bad.

For me I am looking to hopefully get E6600 before the year ends, and maybe a 7900 GT
as there is no ATI equivalent in Toronto, Canada at this price range.

I typcially don't visit PC Perspective, it's Anandtech, HardOCP, HotHardware, Nvnews.net for me. Though from what they have said, they were being diplomatic and that Conroe is looking to be the better deal and not quoting specific figures.
 
Well...any AMD fan remembers when PCPer used to be "amdmb.com" :)

I don't quite remember the time of the change over, but I have a feeling it was definitely during the Northwood C's period, haha.
 
Mav451 said:
Well...any AMD fan remembers when PCPer used to be "amdmb.com" :)

I don't quite remember the time of the change over, but I have a feeling it was definitely during the Northwood C's period, haha.
Click to expand...

This is for you and Coldpower27. They still lean towards AMD and all four of their leaderboard systems are AMD based systems. But these same guys did do Dollar to time saved ratios for Video type apps and Frames Per Second to Dollar ratios for Games.

Ryan Shrout is a lot less AMD biased than Anand or most of the folks who works at Tom's hardware. Folks like Wes Fink who writes at Anand can make Anand's site look way more AMD biased than it is, same for Bert and others at Tom's. AMDmb wasn't as bad as the AMDZone, Van's hardware, Sudhian and etc...IMHO. But, if he, Anand and others who lean towards AMD were impressed by what they saw of Conroe, then that impresses me even more. They'd not talk up Intel when they haven't in the past. Again that's just my opinion. I too like [H] the best because they've been consistant for years and that goes back to the 1133MHz P3 recall. :)

[H] also stood up to the AMD's FUD machine when it came to Hyperthreading. Many Intel users knew it worked while AMD called it a Gimmick and said it hurt performance. In some cases it did, many others it helpped. [H] could have jumpped on the misinformation wagon but that's not what this site is about. I'm not kissing up when I say, THANKS [H]!
 
Donnie27 said:
This is for you and Coldpower27. They still lean towards AMD and all four of their leaderboard systems are AMD based systems. But these same guys did do Dollar to time saved ratios for Video type apps and Frames Per Second to Dollar ratios for Games.

Ryan Shrout is a lot less AMD biased than Anand or most of the folks who works at Tom's hardware. Folks like Wes Fink who writes at Anand can make Anand's site look way more AMD biased than it is, same for Bert and others at Tom's. AMDmb wasn't as bad as the AMDZone, Van's hardware, Sudhian and etc...IMHO. But, if he, Anand and others who lean towards AMD were impressed by what they saw of Conroe, then that impresses me even more. They'd not talk up Intel when they haven't in the past. Again that's just my opinion. I too like [H] the best because they've been consistant for years and that goes back to the 1133MHz P3 recall. :)

[H] also stood up to the AMD's FUD machine when it came to Hyperthreading. Many Intel users knew it worked while AMD called it a Gimmick and said it hurt performance. In some cases it did, many others it helpped. [H] could have jumpped on the misinformation wagon but that's not what this site is about. I'm not kissing up when I say, THANKS [H]!
Click to expand...

I do agree that most enthusiast sites tend to lean AMD's way. But Toms is the furthest thing from being AMD biased. The guys over there bash AMD every chance they get. All in all I think most sites are leaned towards AMD. But I think there is a reason for it. Not that AMD is better then Intel, becouse they certainly are not, but becouse they manage to compete. The fact that AMD is one fifth the size of Intel, and manage to stay competitive is impressive, and they deserve credit for that.

i DO agree tho that Intel does deserve more credit then they got for Hyperthreading. If Intel had implemented it properly, then it could've been the single largest boost in performance since the 286. After all, lets face the truth. ILP is almost at the limits of what can be done today. Intel has done a great job at over extending it with micro, and macro ops fusion, but TLP has alot more potential. DSMT is the way to do it. ALOT more efficiant then adding symmetric cores. The same benefit but with much less hardware. However they did not, so they ALSO deserve alot of critisizm too. The same can be said of AMD, becouse they never even attempted it. At least Intel did.

I also Agree that Hardocp has done a great job at being as nuetral as possible. It is great to get a good look at both sides from one place.
 
Donnie - most Intel fans always cite Tom's as their mainstay for "benchmarks". Remember when the Barton 3200+ came out? Tom's put out an article titled "Paper Tiger - the 3200+ Barton"...or something very similar

Tom's has never, will never be AMD biased. That place has deep roots in its Intel bias, beyond a doubt.
 
duby229 said:
I do agree that most enthusiast sites tend to lean AMD's way. But Toms is the furthest thing from being AMD biased. The guys over there bash AMD every chance they get. All in all I think most sites are leaned towards AMD. But I think there is a reason for it. Not that AMD is better then Intel, becouse they certainly are not, but becouse they manage to compete. The fact that AMD is one fifth the size of Intel, and manage to stay competitive is impressive, and they deserve credit for that.

i DO agree tho that Intel does deserve more credit then they got for Hyperthreading. If Intel had implemented it properly, then it could've been the single largest boost in performance since the 286. After all, lets face the truth. ILP is almost at the limits of what can be done today. Intel has done a great job at over extending it with micro, and macro ops fusion, but TLP has alot more potential. DSMT is the way to do it. ALOT more efficiant then adding symmetric cores. The same benefit but with much less hardware. However they did not, so they ALSO deserve alot of critisizm too. The same can be said of AMD, becouse they never even attempted it. At least Intel did.

I also Agree that Hardocp has done a great job at being as nuetral as possible. It is great to get a good look at both sides from one place.
Click to expand...

Look again? Bert and Frank are AMD biased=P These are the same two who said they'd not want to overclock a Northwood C because they didn't want to make the P4 unstable. They were laughed off the Web. Dr. Tom himself came back and did and overclock updated to the review. Again, it's not the head guy at many sites who does most of the reviews. Van Smith wrote for Tom's and his "A Titan Falls" article is in the AMD Fanb0y Hall of Fame. Same way with Anandtech and Wes Fink.

I'd sound like I was kissing up to Kyle and the gang if I compared them to Tom's or the rest. I don't think they're perfect but they're fairer. None of the mainline Websites dove into overclocking the P4 like [H] and some even repeated AMD's Bogus PR or FUD like Hyperthreading doesn't work while [H] was showing it did.

Most Webmasters seem to follow and cater to their Hits. They do a Poll and if 70% of the hits are from AMD users, many just tilt their reviews that way. I know crap loads of Intel users and folks who use both who don't even bother to post on Message boards. These folks just say why bother?
 
Mav451 said:
Donnie - most Intel fans always cite Tom's as their mainstay for "benchmarks". Remember when the Barton 3200+ came out? Tom's put out an article titled "Paper Tiger - the 3200+ Barton"...or something very similar

Tom's has never, will never be AMD biased. That place has deep roots in its Intel bias, beyond a doubt.
Click to expand...

It's not just benchmarks though. It's how the system is setup, what software is tested and their opinions that matter the most.

I use to belong to Tom's old Delphi forum 6 years ago. I have fond memories of AJ, PopeGoldX and others LOL! Tom himself, other than being Mr. Ego, isn't about to bash Intel like his writers. But Tom's, just like many other tested the 3200+ and said it performed more like AMD's older 2800 or the Intel 2.8GHz. Sure they didn't do like Extremetech and Run 9 Games and 2 other apps, only two apps being Optimized.
 
Well just talked to Intel Engineer up in Oregon. They can't say much about the Conroe at this time. But what they did say was that this is the next big leap. IE: 486-Pentium.
 
rgold28 said:
Well just talked to Intel Engineer up in Oregon. They can't say much about the Conroe at this time. But what they did say was that this is the next big leap. IE: 486-Pentium.
Click to expand...
I can agree with that. As far as power consumption and work per clock cycle, Conroe seems to be a big leap forward.
 
I imagine big improvements are in the works, especially for mobile and server platforms.
 
RancidWAnnaRIot said:
I'd just like to stress.. that a bigger cache doesn't mean you will perform better
Click to expand...

But the way it uses that bigger cache does mean better performance.
 
Donnie27 said:
But the way it uses that bigger cache does mean better performance.
Click to expand...


well.. the "way" a cache is implemented does matter.. but the determining factor of whether you perform better or not is in the actual code being runned..

But, i'm sure intel does massive amounts of cache simulations running different type of benchmarks and such to determine like an optimal cache configuration for across the board gain.. as opposed to gains in some types of a programs, and loses in others.. so in the end to the average user.. there will be a performancing gain...

but i guess this is just getting into garbage that isn't really important..
 
Status
Not open for further replies.
Back
Top