Sign in with
Sign up | Sign in
Your question

P4 2.8 Comments (especially regarding voltage)

Tags:
Last response: in CPUs
Share
August 27, 2002 1:01:43 AM

Ok, i know there are already 2 threads about this topic (which a childish debate over whose post was the first to go through the incredibly slow THG forum server), but here's another. :) 

First, i want to address the whining over voltage. People are moaning and griping about the fact that Intel has raised core voltages in the 2.6, 2.66, 2.50, and 2.80 to 1.525 volts (up from 1.5 volts). Even the THG article about the 2.8 Ghz mentions it (almost as a sarcastic punch to the 2.8 chip). First of all, can I say shut up? I think most of those comments simply comes from miffed fanboys who are trying to deny that the 2.8 beats the 2600. The 1.525V is still lower than the 2600 voltage of 1.65V and has come down a lot since the p4 1.7, 1.8, etc. voltage of 1.7V (or 1.75V). So quit the whining, it's getting old. Both companies have to try to reduce power consumption increases with each new generation, true (the real barrier to Moore's Law - it's not how small you can make the parts, but how hot they will run at and how much power they will consume that is the big problem), but i don't see a reason to jump and point and scream at a .025 increase :)  2.53 -> 2.8 is 10.7% speed increase (Mhz wise, not performance wise, i know), and .025 Volts is only a 1.67% increase. Not too shabby I suppose.

Anyway, once again, I'd like to say Tom's a little biased. Can everyone else see how the 2.8 simply whomps on the 2600? I mean, I'm not reading these benchmarks wrong, am I? It seems that when AMD comes out with something that beats the top Intel chip (the 2600 was and is still comparable to the 2.53, not really a true winner), this guide is first to applaud AMD and comment on (without mentioning anything regarding price, memory, etc.) that it's just plain and simple the fastest processor you can buy. But when flipped (for instance...today), it's hesitently said that the Intel proc is the best, the king, the crown holder. Tom undercuts the win by mentioning the upcoming release date of the 2800 and price vs. performance (or even by pointing out that Intel uses faster RDRAM, where AMD uses DDR - AMD's crutch, not Intel's unfair business decision, lol). If you're going to award speed crowns, let it just be speed crowns. If you're going to be a guide to tell consumers which chip is best to buy, then do more research on that aspect (instead of so many quake3 and Sandra charts, show some charts of pricing at different places online and in stores). Consistency in doing these reviews is the most important thing, IMO. Granted, a lot of us care about video card speeds, hard drive capabilities, and those cool looking computers the size of toasters, but isn't the most looked at aspect of THG the chip review section? I think it is...

Finally, can i say that the 2600 release was a little sneaky. *looks at pricewatch*...where are the 2400 and 2600's? *looks harder* I don't know. I can't see them :) 
But the 2.8 is there, even if on paper it was released a week later than the 2600. I wonder when a consumer will actually be able to see a fabled 2600? AMD's site lists "world-wide availability" at end of september. So i wonder...will they say they have released the Hammer in q1, 03, which then means they made ONE and it can only be tested internally, and then full availability will be in 04? Honestly...if they really were trying to just hit mid-market, they'd market their chips more honestly (that almost sounds like a palindrome, lol).

There's my uh....4 cents :) 

Athlons and Pentiums are just melted rock. Who’s rock is better? Who cares, let’s play some games

More about : comments voltage

August 27, 2002 1:16:46 AM

Mr. Worf, flameshields up. Arm anti-fanboy torpedoes. Charge forward fact-banks.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 27, 2002 2:21:16 AM

I dont mind a .025 v increase at all.... whats so bad about it?

real philosophy of life: "do onto others what you dont want them do onto you"
Related resources
August 27, 2002 2:48:02 AM

there a part that i dont: RDRAM do not overclocke well and the new 533 fsb dont not overclock well too.... but he have use it for is 3.7 ghz for 3Dmark.


When they increase voltage it because it wont be able to produce a large amount of Cpu at this speed so rainsing volatge help in the chip yield.The same will happen with prescott when they will move to 1.2 to 1.2025.Also the new stepping look like a better overclocker

At the end i have speak with a horny lady<P ID="edit"><FONT SIZE=-1><EM>Edited by juin on 08/26/02 10:49 PM.</EM></FONT></P>
August 27, 2002 2:50:28 AM

If people thought I was pro-AMD because I don't give praises to Intel enough, then here's one:
Your point is right wolverino, whining about this is so idiotic. I am amazed as hell that the Intel chips are at 2.8GHZ and STILL and ONLY use about 1.5V. This to me shows great work in the architecture (unless it's another pipeline length reason?). The fact at 2.8GHZ not only you got a winner CPU, BUT it's still cool. You can run it with a 2000RPM fan and it'd still not peak 50ºC. I think this is worth giving major praise, because AMDs at 2.13GHZ use 1.65V and near 1.70V once more.

And I didn't do this to cool the pro-AMD attacks at me, but it is my personal opinion straight up.

--
And now, an advice from your friendly Nike shoes slogan: JUST DO HER!
August 27, 2002 6:39:08 AM

The reason THG might seem slightly biased towards AMD is because Intel has ruled for years and years and years and finally someone can truly challenge them. It's exciting to see competition in a market that was devoid of it for so long.

AMD 1800+
GeForce 4 MX 440
512MB DDR
Windows 2000
Soundblaster Live! 5.1
52x CD-ROM
August 27, 2002 7:02:44 AM

IMO P4 2.8 beats at all and, without having done and extensive research and comparison, I feel also AMD rating is starting to point to high. AMD's 66Mhz extra move points in that direction, and I hope they will do it again in future CPU in order to stay "more or less" at the level they say.

Also I think that the configuration information given in the article is useful, but some prices about it could be even better. What about system 1 : $xxxx and system 2 : $xxx. I know we are looking for the fastest system no matter how much cost, but I would appreciate this information.

DIY: read, buy, test, learn, reward yourself!
August 27, 2002 7:11:17 AM

That's kind of what i'm beginning to think. I mean everyone likes an underdog, likes to root for the underdog and see the big guy get challenged (how many people hate microsoft?). I just think that after awhile it gets old. But then again, everyone has their own biases, right? :)  "Objective news" is really the largest oxymoron.

Athlons and Pentiums are just melted rock. Who’s rock is better? Who cares, let’s play some games
August 27, 2002 9:36:07 AM

Is it possible we have a case of Intel "fanboys" getting annoyed? Remember, it's a processor, not your life...although...;)

AMD Is An Anagram Of MAD, Intel Is An Anagram Of INLET, Cyrix...Ah Who Cares?
August 27, 2002 1:39:58 PM

Quote:
Is it possible we have a case of Intel "fanboys" getting annoyed?

I don't think that's the case at all. Here's the difference:

<font color=blue>Intel does a paper launch, AMD 'fanboys' heckle Intel.</font color=blue>
<font color=green>AMD does a paper launch ... general silence.</font color=green>

<font color=green>AMD launches fastest CPU, praises are sung loudly and for weeks.</font color=green>
<font color=blue>Intel launches fastest CPU, constant complaints that belittle Intel's accomplishment.</font color=blue>

I don't think it has <i>anything</i> to do with Intel 'fanboys' getting annoyed. It is much more about the hypocracy growing in AMD and AMD 'fanboys'. Plenty of people who are even AMD fans are starting to notice this, and certainly the 'middle ground' people (who just care about the end product and not the name stamped on it) are noticing it and commenting on it.

Hell, if anything it is a testament to Intel 'fanboys' that there <i>haven't</i> been a massive number of taunting threads or insults at AMD over the AXP 2600+ release and the subsequent P4 2.8 release.

Even when the Intel 'fanboys' have perfectly good reasons to mock and laugh at AMD, they aren't doing so. I think that says a lot.

Quote:
Remember, it's a processor, not your life...although...;)

No more than a car is your life, or a house is your life, or a diamond and gold wedding ring is your life... It's about a commitment of a considerable sum of time and money to invest in your future. Intelligent people research what meets their needs best. Frankly, it holds a lot more import on many people's lives than what cut of steak they had for dinner or what cereal they poured into the bowl for breakfast.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 1:50:01 PM

Quote:
I mean everyone likes an underdog, likes to root for the underdog and see the big guy get challenged

It reminds me of the saying "You either lead, follow, or get out of the way." I always heard that and would say to myself, "What do I care what other people are doing?" I don't lead. I don't follow. And I don't get out of the way either. I just simply do what suits me best, end of story.

The same is true for the whole root for the underdog and complain about the evil overlord thing. Honestly, I don't care who's the underdog and who's the evil overlord. All that I care about is what product best meets my needs. End of story.

Maybe it makes me a boring person. **really big shrug** Beats me. All I know is I couldn't care less who's name is on the products I use. All I care about is how that product performs.

Quote:
(how many people hate microsoft?)

I vastly dislike Microsoft for a considerable number of reasons, none of which have to do with them being a 'monopoly. But then, I also have a number of qualms with Linux and the Linux community as well. I can make cases for and against open source. In the end, <i>nothing</i> is perfect. My eyes see no knight in shining armor fighting the demon of evil incarnate. All I see is chaos and disorder caused by people with ideals who can't actually execute them effectively.

Ugh. Why am I feeling so cynical today?

Oh yeah, it's because I <i>am</i> cynical. ;) 

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 3:16:01 PM

Cudos to Wolverine for making another thread. Those kiddies whine too much in the other ones.

I think the thing with AMD is yes they are the underdog, and everone loves an underdog. But I think another big reason why there are so many AMD fans out there is becaues of ignorance. YES, there are the true amd fanboys that live and die by AMD. But there are also those end users who don't really know much about computers and don't really udnerstand much about processors etc and they hear things and read things from placed like THG and say "oh yeah AMD rocks man! Intel sucks!" But they have no clue WHY at all. I have a friend who I was talking to a week ago who said he thinks he's going to go AMD for his next computer. I ask him "why?" he says "I dunno, I just think it's better". Why? No particular reason. A lot of less knowledgeable people just follow what the general consensus has been which has been AMD for a while. Maybe that'll change now that Intel hs pushed itself a far distance away from AMD.

This is exactly why I Think it's so important that THG and other hardware review sites try and be as unbiased as possible and just report the facts rather than personal opinions. It's these small little person grudges that swing people from AMD to Intel and Intel to AMD. THG should be reportiung the facts and statistics and let the end user decide what to choose.

Slvr, you're right though.. Doesn't really seem like Intel fanboys sneer as much as AMD ones. None of these Intel threads have people laughing at AMD etc. But as soon as the 2600+ came out on paper al these people start flaming Intel (god knows why since the 2.53 still performs better).

It's like a typical high school. AMD are the cool jocks and Intel people are the nerds. Whos going to get all the chicks and money in the end?

Cooj

I am ignorant no more!
August 27, 2002 4:59:19 PM

Thank you slvr, for better explaining what i was trying to say. I just am tired of hypocritcal behavior on the part of technical review websites. I love AMD and how they've made Intel uncomfortable and forced them to produce better perfomance at a cheaper price (top end pentiums are being released now sub-600 where as they used to be easily 1000+). AMD does constantly surprise people by being able to push out new products (and fufill demand...most of the time) with a lot less resources than Intel. However, constantly trying to degrade Intel so that AMD can be seen as better than it really is and snag more customers is frankly disgusting to me.

Good point about the life thing. No this isn't our lives, but computers are interesting and important to us (we wouldn't be here otherwise). We are spectators to a fight of the future (Tom constantly includes benchmarks from heavily overclocked top-end chips, speeds that won't really hit the market officially until many months from now) and it's exciting, IMO. I personally am most interested in where these little chips are going to start going. When a top end pentium starts being released at 200 dollars, what are we going to do we that incredible available and strong computing power? That's where this race is leading (which is why i think flash memory is now usually a half step ahead in terms of shrinks - it's very useful for direct consumer electronics). Anyway...time to call my g/f :) 

Athlons and Pentiums are just melted rock. Who’s rock is better? Who cares, let’s play some games
a b à CPUs
August 27, 2002 5:12:26 PM

A lot of people would point to the measly 0.025v core increase and call it factory overclocking. Well, Intel has a habit of releasing UNDERVOLTAGE CPU's. I had a Celeron 566 that ran at 1.50v I had a PIII 700E that ran at 1.65v. These came off the same production line at similar times with the same core revision. Was the 700E overvolted? No, the 566 was undervolted. Intel does this to maintain the most heat efficiency they can at a given clock rate.

In fact, when I pushed the voltage of the 566 up to 1.75v, people FREAKED OUT and said I would burn the CPU. Even Tom said it was a bad idea to go more than 10% over standard voltage. But when I did it with the 700E, everybody seemed OK with that. Even Intel released 1.75v versions.

So the answer here is not "what was the original voltage" but instead "What are the ideal voltages for the highest and lowest clocks". For the NW, anything up to 1.65v seems to only increase heat marginally, from 1.65v to 1.75v the rate of heat increase seems to be intermediate, and from 1.75v up, high. So as long as Intel isn't going about 1.65v, who's to say that the former speeds were not simply "undervolted" for improved efficiency? Not I!

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
August 27, 2002 5:29:10 PM

Personally, that AMD or Intel is the chip maker makes little difference to me. The fact that at the time I was looking for a new CPU 2+ years ago and bought an AMD Athlon had more to do with performance and value than name. Today, I am looking at upgrading my CPU and most likely will purchase a P4. Why? Performance and Value.

With the exception of the "latest and greatest" that are being popped out on both sides, the cost varience is nominal until you get to the 'older' chips where AMD has the value edge. Considering that the P4 2.26 performs as good and a little better than the XP 2200+ the matter that there is a nominal cost difference makes the P4 a better buy. I also like the built in thermal protection on the P4s. Since I want a chip that will last another 2+ years, to me the P4 like the most attractive chip at this point in time.
August 27, 2002 6:04:38 PM

Intel use 1.3 volt for mobile P4 wich made much coller even so they hit 2.0 ghz easy.Intel have a PDF show that using less voltage like 1.6 to 1.5 will only decrease the max speed by 200 mghz but will decrease the heat output by a large amount.

At the end i have speak with a horny lady
August 27, 2002 6:17:26 PM

That's a good point Crashman about the undervolting. Intel has always been good at minimizing their voltage.

I just wish Intel would have done more on the Ultra Low Voltage Mobile Celerons. Looking at that chip makes one see how Intel could totally crush VIA's 'low power/low heat' C3s. I mean honestly, a 650MHz Celeron that takes only 1.1V of juice and puts out a whole 7W of heat...

Imagine what Intel could do with a low-power chip like that at 0.09 microns! You'd probably be able to put the thing into a PDA. Heh heh.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 7:03:52 PM

I agree, the whining should stop. I mean, it's a tiny increase in voltage, and it was needed to supply the P4 with enough power. ALso, about the whole argument with the p4 3Ghz needing a new mobo, that's because all current NW mobo's are desgined to supply <i> about</i> 60 watts of power. The 3ghz NW will need more than 60 watts, or else there might be stability problems, and/or voltage problems. Also, considering the high expectations that Intel puts on it's chipsets and CPU's, they are NOT going to stop the release of the new mobos, or rework the P4, just to lower the voltage. Rumour has it, the new mobo's will have a <i> slightly</i> different socket, and could be compatible with Prescott.

Although I'm being quiet about it, it really angers me that AMD did a paper launch of the XP 2600+. The fact is, it's not listed on any of the sites I've visited including newegg, or pricewatch. But the 2.8 P4 is listed on both sites. The truth is, you won't be able to buy the XP 2600+ for a few weeks, and the official price of the 2.53B is <b>lower</b> than the official 2600+ price. On newegg, the 2.5 is listed at $280, while on pricewatch it's listed at $249. EVEN IF RDRAM is more expensive than DDR, if I needed a new computer, I would definitely get the 2.5 right now. But, I don't need a new one, since my 1ghz pIII is just fine for me.

Slvr has hit the nail on the head. Even though the 2.8 is whopping the 2600+, and the 2.5B is not only equal, but costs lower officially, and you can buy it right now, as opposed to waiting for the "mystery date" of when the 2600+ is coming. With all that in mind, I'm being pretty quiet about this whole argument.

and slvr, Intel's reserved strongARM CPU's for PDA's. Believe or not, I'm EAGERLY awaiting the Banias. When it comes out at the beginning of 2003, it will be a silent monster. I'm really excited about it. I haven't been this excited about a CPU ever since NW came out. Banias will output <b> 25W of heat at 1.6Ghz</b>. How's that for cool? And it is being said that it has the ability to <b>outperform</b> a 2.4B Northwood! When I heard this, I was like OMG, this thing is gonna make Tbred look really bad, and will seriously put a dent in the PR rating. It will only be a mobile chip (there might be a desktop version in the future) and it will use an updated version of Speedstep. Intel has stated itself that it will outperform Northwoods clock per clock. I read this in an IDF article somewhere. Also, Intel said that the 1.6 will have a vcore of 1.35 (25W heat output), but can speedstep down to 0.85, and if the heat output lowers in a linear pattern, then the heat output will be about 15W at 0.85V.

Anyways, I'll stop the off-topic rambling now.

- - - - -
Montecito - successor to Madison, and one monster of a CPU.
August 27, 2002 7:29:49 PM

Quote:
and slvr, Intel's reserved strongARM CPU's for PDA's. Believe or not, I'm EAGERLY awaiting the Banias. When it comes out at the beginning of 2003, it will be a silent monster. I'm really excited about it. I haven't been this excited about a CPU ever since NW came out. Banias will output 25W of heat at 1.6Ghz. How's that for cool? And it is being said that it has the ability to outperform a 2.4B Northwood! When I heard this, I was like OMG, this thing is gonna make Tbred look really bad, and will seriously put a dent in the PR rating. It will only be a mobile chip (there might be a desktop version in the future) and it will use an updated version of Speedstep. Intel has stated itself that it will outperform Northwoods clock per clock. I read this in an IDF article somewhere. Also, Intel said that the 1.6 will have a vcore of 1.35 (25W heat output), but can speedstep down to 0.85, and if the heat output lowers in a linear pattern, then the heat output will be about 15W at 0.85V.

Now that IMO is pure hype. Just thinking that a mobile chip will outperform a 2.4 B MODEL, is almost making me laugh.
It'll require more than just a document so I'd beleive this. Hell even Prescott at 1.6GHZ is not supposed to be beating a 2.4B if all goes well!

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
a b à CPUs
August 27, 2002 7:33:29 PM

Hmmm, hardly sound accurate considering the numbers overclockers are seeing.

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
a b à CPUs
August 27, 2002 7:38:46 PM

Hmm, a PDA with virtual 1600x1200 display by glasses, a virtual keyboard, etc. Now that would be a good portable system!

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
August 27, 2002 7:53:38 PM

Ya I know, that's why I was shocked when I read it. Unfortunately, I also doubt that it's true.

- - - - -
Montecito - successor to Madison, and one monster of a CPU. <P ID="edit"><FONT SIZE=-1><EM>Edited by Dark_Archonis on 08/27/02 03:55 PM.</EM></FONT></P>
August 27, 2002 8:04:03 PM

You scared me for a while dude! :tongue:

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
August 27, 2002 8:40:02 PM

<b>Dark_Archonis</b>
Quote:
and slvr, Intel's reserved strongARM CPU's for PDA's.

Yeah, I know, but that's my point. It's so ... weak. If Intel just even used 0.09 micron tech to reproduce their Ultra Low Voltage Celeron, it'd be noticably better than a strongARM.

Quote:
Believe or not, I'm EAGERLY awaiting the Banias. When it comes out at the beginning of 2003, it will be a silent monster. I'm really excited about it. I haven't been this excited about a CPU ever since NW came out. Banias will output 25W of heat at 1.6Ghz. How's that for cool?

Sounds good. It'll make a fine notebook. :) 

<b>Eden</b>
Quote:
Now that IMO is pure hype. Just thinking that a mobile chip will outperform a 2.4 B MODEL, is almost making me laugh.
It'll require more than just a document so I'd beleive this. Hell even Prescott at 1.6GHZ is not supposed to be beating a 2.4B if all goes well!

I agree in that it such performance sounds like a lot of hype. However, such is not impossible, just improbable.

If Intel were to extend and expand the P3 core using what they learned developing the P4 and using 0.09 micron technology, it could be entirely possible for a 1.6GHz chip to have a vastly higher IPC than a Northwood and possibly even perform as well as the hype suggests.

<b>Crashman</b>
Quote:
Hmm, a PDA with virtual 1600x1200 display by glasses, a virtual keyboard, etc. Now that would be a good portable system!

That would be pretty sweet. However that might scare a few people to see me working on my novel with one of those. Heh heh.

I'd even settle for a slightly larger PDA with a 650MHz P3, 128MB RAM, 640x480 24-bit portrait-layout-oriented display, a foldable keyboard expansion (not built in so that keyboard can be easily replaced should it become damaged or sticky), a 1GB IBM Microdrive, and a free expansion slot if it could run a <i>normal</i> version of Windows (preferably 98SE, though 95b would work just as well and save HD space) and MS Office.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 11:54:31 PM

Quote:
I agree in that it such performance sounds like a lot of hype. However, such is not impossible, just improbable.

If Intel were to extend and expand the P3 core using what they learned developing the P4 and using 0.09 micron technology, it could be entirely possible for a 1.6GHz chip to have a vastly higher IPC than a Northwood and possibly even perform as well as the hype suggests.

Then lemme ask you this: Would INTEL really do that?
Notice I emphacize the name and honorable company, INTEL! :wink:

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
!