Tom's P4 Overclocking Review

[Jeff7181]

Originally posted by: Zebo

Originally posted by: Jeff7181
I believe by "again" he was referring to Socket 775 BEFORE the overclock lock could be bypassed. Meaning... before, the Socket 775 platform was limited in what you could do with it so it wasn't a very good choice for overclockers. Now that the lock can be bypassed, the Socket 755 Pentium 4's is again a valid choice for overclockers. It's powerful in it's overclocking ability, thanks to this new motherboard.

astute observation...some 50 posts later.

Yet people are still arguing about 3.6 GHz vs. 3.72 GHz :roll:

 

[carlosd]

Originally posted by: Elcs

Originally posted by: carlosd
Where the hell are the A64 CPU's in this review. Is intel afraid?
When you launch a new product you compare it with the best, and the most powerfull processors are Athlons 64. You don't know nothing about testing. I am an electronics enginneers, and you don't compare your new developments with your own products, it is stupid and meanigless technically, but form the point of view of the marketing it is a smart move.

Did you even read any part of that review? It doesnt sound like it.

ASUS released a BIOS which allows certain multiplier tweaks on certain Prescott cores. They werent showing off a new Intel processor, just using new features to overclock old ones.

It is the same or even worse. They are comparing new development (a new BIOS is a new development son) with the same plataforms.

 

[gururu]

I think its an awesome review. prices on these chips are dropping fast, and Asus' solution will breathe new life into them. there was no need to even bring up AMD. whether they have a faster solution or not at a similar price point is not the point of the article.

 

[Avalon]

Originally posted by: Jeff7181

Originally posted by: Zebo

Originally posted by: Jeff7181
I believe by "again" he was referring to Socket 775 BEFORE the overclock lock could be bypassed. Meaning... before, the Socket 775 platform was limited in what you could do with it so it wasn't a very good choice for overclockers. Now that the lock can be bypassed, the Socket 755 Pentium 4's is again a valid choice for overclockers. It's powerful in it's overclocking ability, thanks to this new motherboard.

astute observation...some 50 posts later.

Yet people are still arguing about 3.6 GHz vs. 3.72 GHz :roll:

Well, you must admit that the wording of this part of the article isn't the best:

The benchmarks are self-explanatory. With DDR2-710, FSB1066 and a 3.72-GHz CPU clock speed, the P4 without a doubt becomes very powerful again.

I can understand why a lot of people take it to mean that this has all sorts of implications, such as that the new chips overclocked are powerful compared to A64, or more powerful compared to non 775 P4s. But, you do make a good point. It does make more sense to take that as the new P4s on the Asus board overclocked are a very powerful overclocking option again. I'm checking back on the charts, though, and I would have thought that the extra 400mhz, memory bandwidth, and faster FSB would have made more of a difference in some of those tests over the stock 3.6ghz 775 P4. Ah well, not my problem

 

[Zebo]

Originally posted by: Lithan
Yeah. So he's saying 3.6ghz*1.15 is slow, 3.72ghz is very powerful. Sounds like a load of bull to me.

No he's saying now that the lock has been bypassed it makes the socket/chipset a powerful option again due to overclcoking. Forget the numbers, it's fully unlocked now which before was'nt, and enables full potential of chip. "Powerful again" in the sense 865/875 was powerful in the right hands.

 

[Duvie]

Exactly Zebo.....Haven't we already seen then time and time agin anyways??? The P4 with a lower multiplier will allow higher fsb and therefore the P4 will benefit from the increased bandwidth and maybe even more so with the newer chipset.

Like oldfart has stated it doesn't always have to be Intel vs AMD in every review. This wasn't cpu shootout or showcase it was more about a motherboard hardware review and how the tweaks of the new bios can give it an edge with the p4 chip. More so against its other competitors making the same p4 chipset.


YOu AMD ppl have the performance crown....Enjoy it for whatever it is worth to you just don't expect everyone to bow down and lick your a** everytime anything Intel or non amd is reviewed.....

 

[Viditor]

Originally posted by: Duvie
Exactly Zebo.....Haven't we already seen then time and time agin anyways??? The P4 with a lower multiplier will allow higher fsb and therefore the P4 will benefit from the increased bandwidth and maybe even more so with the newer chipset.

Like oldfart has stated it doesn't always have to be Intel vs AMD in every review. This wasn't cpu shootout or showcase it was more about a motherboard hardware review and how the tweaks of the new bios can give it an edge with the p4 chip. More so against its other competitors making the same p4 chipset.


YOu AMD ppl have the performance crown....Enjoy it for whatever it is worth to you just don't expect everyone to bow down and lick your a** everytime anything Intel or non amd is reviewed.....

It's THG...the choice of words was spun and poorly chosen, but that's to be expected from them. Other than that, I have no disagreements.

 

[hytek369]

hmm, interesting article. i didn't know intel locked their cpus but not surprised

 

[clarkey01]

"YOu AMD ppl have the performance crown....Enjoy it for whatever it is worth to you just don't expect everyone to bow down and lick your a** everytime anything Intel or non amd is reviewed..... "

Duive, theres no need for that, agree or not, dont be making commnets like that, end of the day, its only CPU'S were talking, not life or death. You bitter man. And when theres a point to be made on my side I'll have a little fun, I dont give my opinions out trying to get a " Hail Clarkey01", there just for fun, to help and my views.

 

[imported_SLIM]

Did anybody else notice that he used engineering sample CPUs? I thought they were factory unlocked. It seems like Anand has done reviews before where he under/overclocked engineering sample P4s by changing their multiplier for comparison purposes.

SLIM

 

[Drayvn]

Originally posted by: Viditor

Originally posted by: Duvie
Exactly Zebo.....Haven't we already seen then time and time agin anyways??? The P4 with a lower multiplier will allow higher fsb and therefore the P4 will benefit from the increased bandwidth and maybe even more so with the newer chipset.

Like oldfart has stated it doesn't always have to be Intel vs AMD in every review. This wasn't cpu shootout or showcase it was more about a motherboard hardware review and how the tweaks of the new bios can give it an edge with the p4 chip. More so against its other competitors making the same p4 chipset.


YOu AMD ppl have the performance crown....Enjoy it for whatever it is worth to you just don't expect everyone to bow down and lick your a** everytime anything Intel or non amd is reviewed.....

It's THG...the choice of words was spun and poorly chosen, but that's to be expected from them. Other than that, I have no disagreements.

Yup, Tom made a bad choice on words it seems.

 

[Jeff7181]

Originally posted by: Avalon

Originally posted by: Jeff7181

Originally posted by: Zebo

Originally posted by: Jeff7181
I believe by "again" he was referring to Socket 775 BEFORE the overclock lock could be bypassed. Meaning... before, the Socket 775 platform was limited in what you could do with it so it wasn't a very good choice for overclockers. Now that the lock can be bypassed, the Socket 755 Pentium 4's is again a valid choice for overclockers. It's powerful in it's overclocking ability, thanks to this new motherboard.

astute observation...some 50 posts later.

Yet people are still arguing about 3.6 GHz vs. 3.72 GHz :roll:

Well, you must admit that the wording of this part of the article isn't the best:

The benchmarks are self-explanatory. With DDR2-710, FSB1066 and a 3.72-GHz CPU clock speed, the P4 without a doubt becomes very powerful again.

I can understand why a lot of people take it to mean that this has all sorts of implications, such as that the new chips overclocked are powerful compared to A64, or more powerful compared to non 775 P4s. But, you do make a good point. It does make more sense to take that as the new P4s on the Asus board overclocked are a very powerful overclocking option again. I'm checking back on the charts, though, and I would have thought that the extra 400mhz, memory bandwidth, and faster FSB would have made more of a difference in some of those tests over the stock 3.6ghz 775 P4. Ah well, not my problem

If people didn't go into the article with the mindset that it was a pro-Intel/anti-AMD article to begin with, and realize that the article was about a motherboard and chipset, they would have realized that he was referring to the Pentium 4 Socket 775 line with a motherboard that bypasses the overclock lock as opposed to the Pentium 4 Socket 775 line with a motherboard that doesn't bypass the overclock lock.

It's a matter of perception. Those of you that accuse Tom of being Intel bias and complain about this comment also seem to have an AMD bias. So you're really no better than Tom... in fact you're worse cause you don't have your own website

 

[DAPUNISHER]

It's a matter of perception. Those of you that accuse Tom of being Intel bias and complain about this comment also seem to have an AMD bias. So you're really no better than Tom... in fact you're worse cause you don't have your own website

And Jeff lays the smackdown :beer: As to the matter of perception comment, I agree completely. Or as Obi Wan told Luke "Luke, you?ll find that many of the truths we cling to depend greatly upon our own point of view." :light:

 

[jiffylube1024]

Originally posted by: hytek369
hmm, interesting article. i didn't know intel locked their cpus but not surprised

What are you talking about? Intel has always locked their CPU's multipliers, back to the Pentium 1 days.

Now they're also enabling some sort of 'limit' in the chips which theoretically allowed a ~15% (maximum) FSB overclock before they screw up. Fortunately, it seems ASUS (and others to follow) have found a way around this new obstacle.

 

[oldfart]

Originally posted by: jiffylube1024

Originally posted by: hytek369
hmm, interesting article. i didn't know intel locked their cpus but not surprised

What are you talking about? Intel has always locked their CPU's multipliers, back to the Pentium 1 days.

Now they're also enabling some sort of 'limit' in the chips which theoretically allowed a ~15% (maximum) FSB overclock before they screw up. Fortunately, it seems ASUS (and others to follow) have found a way around this new obstacle.

I think I had one of the last ever unlocked Intel CPU's. It was the PII 333. The default was 5 x 66 = 333. You couldn't go over 5x, but you could go lower. I ran it @ 4 x 103 = 412. They changed it half way through the year and locked it @ 5x.

 

[jiffylube1024]

Originally posted by: oldfart

Originally posted by: jiffylube1024

Originally posted by: hytek369
hmm, interesting article. i didn't know intel locked their cpus but not surprised

What are you talking about? Intel has always locked their CPU's multipliers, back to the Pentium 1 days.

Now they're also enabling some sort of 'limit' in the chips which theoretically allowed a ~15% (maximum) FSB overclock before they screw up. Fortunately, it seems ASUS (and others to follow) have found a way around this new obstacle.

I think I had one of the last ever unlocked Intel CPU's. It was the PII 333. The default was 5 x 66 = 333. You couldn't go over 5x, but you could go lower. I ran it @ 4 x 103 = 412. They changed it half way through the year and locked it @ 5x.

Ah, the gold old P2 333. I remember my uncle had one of them, while I was running a G3 266 MHz (the last Mac I'd ever own, thankfully ).

 

[oldfart]

I got it just before the infamous Celeron 300A came out and paid A LOT more for it. I was peeved when I saw everyone getting 450+ Mhz with them!

 

[raildogg]

Originally posted by: oldfart

Originally posted by: clarkey01
Toms Review

No A64 benchmarks lol and he goes " The benchmarks are self-explanatory. With DDR2-710, FSB1066 and a 3.72-GHz CPU clock speed, the P4 without a doubt becomes very powerful again. "

Thats like saying " A 3Ghz celeron becomes very powerfull when paired up to the celeron 266", apples to apples, A64 (3800+/fx 53) with no overclocking would still beat a 4ghz P4 in most test and with a bit of overcloking (2.6Ghz-2.8Ghz) it would wipe the floor with it.


FX 55/4000+ Out 11th of October btw, has Intel even got a 3.8Ghz part out yet ?

Why must everything always be an AMD Vs Intel or nVidia Vs ATi thing? Maybe it was just an article to test overclocking on this particular setup with the new unlock feature? It wasn't meant to compare one brand Vs another. You can do that kind of a review you know.

So, a P4 @ 3.72 GHz is not a powerful CPU?

 

[CaiNaM]

Originally posted by: Duvie
YOu AMD ppl have the performance crown....Enjoy it for whatever it is worth to you just don't expect everyone to bow down and lick your a** everytime anything Intel or non amd is reviewed.....

at least in single tasks, and then it's not by that much. frankly i'm disappointed in my a64's ability to handle 2 cpu intensive tasks simultaneously. intel's HT seems a very good feature that is often overlooked - especially in cpu reviews.

 

[clarkey01]

raildogg, read my next reply and you'll know that I said I agreed and I'd jumped the gun. But other points have been taken up by others, so you might want to focus on issues that are still being addressed.

 

[clarkey01]

CaiNaM, what apps you running ?

My 2000+ and 2.8C act the same, I know the 2.8c is far better, but when I run word, access, mp3's, C++ and MSN etc it runs fine, only a tiny lag, so tiny its funny i notice it, my 2.8C is smooth tho.

 

[imageqst]

Wow.. some of you AMD guys make me laugh.

Anyways, I thought the review was decent, not the best and not worded well at times, but overall it was pretty clear and had a point to it (examining the unlock feature of the asus board).

 

[clarkey01]

"Wow.. some of you AMD guys make me laugh. "


so do some of you Intel guys , unless your not reffering to me, then I take it back.

 

[imageqst]

Originally posted by: clarkey01
"Wow.. some of you AMD guys make me laugh. "


so do some of you Intel guys , unless your not reffering to me, then I take it back.

If I was referring to you, I would have quoted you.

And who says I'm an "Intel guy"? I've owned both and have had bad experienced with both. I'm not more of a fan of either one more than the other really.

I'd be using AMD now if my last AMD didn't kill itself (this was a while back when they ran hot, AMD approved heatsink/fan, good paste, etc, it still died). Currently I do have 3 Intels though (a Pentium 233mhz router/firewall running linux, a P4 1.8 that will be a low end HTPC box and a 3.2 P4) but my next machine will likely be built with an AMD core unless intel does something right.. lately AMD has been looking more appealing to me.

To use another car analogy.. some of the AMD posts in this thread are like Corvette owners, nothing else is good. That doesn't mean that Corvette's are bad, just that sometimes some of their owners can be very close minded.

edit: edited to more clearly make my point.

edit 2: just in case you were wondering, I wasn't upset or mad about your reply, sorry if this reply makes it seem that I was

 

[CaiNaM]

Originally posted by: clarkey01
CaiNaM, what apps you running ?

My 2000+ and 2.8C act the same, I know the 2.8c is far better, but when I run word, access, mp3's, C++ and MSN etc it runs fine, only a tiny lag, so tiny its funny i notice it, my 2.8C is smooth tho.

in your example tho none of the background apps require high cpu priority; try running something where 2 requests priority tho, and the tiny lag will turn into a very obvios one.

the specific reason i notices was from running 2 "windowed" games which request 100% cpu simultaneously.

http://forums.anandtech.com/me...6675&enterthread=y

at first i thought it was something to do with my mb/chipset/bios, but after talking to many ppl over the last week, and reading forum posts, it seems it's just inherent to the a64. i found i can get it to work somewhat satisfactorily by adjusting priorities, but it runs perfectly w/o an issue on my intel platforms (and i can even run mp3s or browse the web while i have 2 clients running).
the strange thing tho is that even my wife's xp 3200+ handles this better (tho not nearly as well as my p4c 3.2) than my a64. part of this may certainly be how efficiently (or rather how unefficently) winxp deals with tasking threads or with the a64 architecture itself, but the end result is still the same.