- Feb 10, 2003
- 2,444
- 0
- 19,780
- Feb 10, 2003
- 2,444
- 0
- 19,780
You mean prescott? No, I'm not stupid. 
You do know that this is not an "AMD is GOOD" or "INTEL is GOOD" issue we're talking about, don't you?
<font color=red><b>M</b></font color=red>ephistopheles
You do know that this is not an "AMD is GOOD" or "INTEL is GOOD" issue we're talking about, don't you?
<font color=red><b>M</b></font color=red>ephistopheles
"Prescott 's enhancements actually give it a steeper increase in performance per increase in clock. Not only can Prescott be clocked higher than Northwood, but as its clock speed is increased, it will start to outperform similarly clocked Northwood CPUs".
Why do they assume that something hidden inside scotty would allow it to defy the basic laws of physics? Is it not much more likely that at speeds of 2.8,and 3.0,the new pipes are not full?
It is also fine and dandy to say any power for perf, but when that power requires more power to remove more heat,we can talk cascade failure. When your 150 watt cpu starts needing a 5hp fan to keep it from incinerating, how will you deal with the noise. When the ambiant temp is already close to 40*c where will you put that heat. No scotty is not great news, now or in the forseeable future.
An advisory to amd fans Watch out for the rats.
Why do they assume that something hidden inside scotty would allow it to defy the basic laws of physics? Is it not much more likely that at speeds of 2.8,and 3.0,the new pipes are not full?
It is also fine and dandy to say any power for perf, but when that power requires more power to remove more heat,we can talk cascade failure. When your 150 watt cpu starts needing a 5hp fan to keep it from incinerating, how will you deal with the noise. When the ambiant temp is already close to 40*c where will you put that heat. No scotty is not great news, now or in the forseeable future.
An advisory to amd fans Watch out for the rats.
POPE_ETERNAL
Distinguished
- Dec 17, 2003
- 14
- 0
- 18,510
see what happens
you look for a savior PRESSHOTT
now you have to make a million excuses why its SLOWER than the previor CPU its replacing
WHeres HARDWARE BOSS ... guess he is working on the same kind of SPIN you are
WHAT IF's and WHEN THEY DO THIS BS
you look for a savior PRESSHOTT
now you have to make a million excuses why its SLOWER than the previor CPU its replacing
WHeres HARDWARE BOSS ... guess he is working on the same kind of SPIN you are
WHAT IF's and WHEN THEY DO THIS BS
kinney
Distinguished
- Sep 24, 2001
- 2,264
- 18
- 19,785
BTW do you have to keep calling him bob?
That is annoying to read (read: snide).. if you continue to do so, what is your first name?
Otherwise I'm sure I can come up with a more creative name for you.
----
Support the NV/AMD/IBM axis of evil.
Who cares about HL2/D3 when we have Call of Duty today!!
That is annoying to read (read: snide).. if you continue to do so, what is your first name?
Otherwise I'm sure I can come up with a more creative name for you.
----
Support the NV/AMD/IBM axis of evil.
Who cares about HL2/D3 when we have Call of Duty today!!
Spitfire_x86
Splendid
- Jun 26, 2002
- 7,220
- 0
- 25,780
Welcome back, popey!
----------------
<b><A HREF="http://geocities.com/spitfire_x86" target="_new">My Website</A></b>
<b><A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new">My Rig & 3DMark score</A></b>
----------------
<b><A HREF="http://geocities.com/spitfire_x86" target="_new">My Website</A></b>
<b><A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new">My Rig & 3DMark score</A></b>
Ok I have some mix feelling.Intel have increase the pipeligne how many time they will do that.I will guess that there stage are unbalance and would new a big rework.It may take about 6 to 9 month before prescott become a platform of choice.SSE3 socket T .......
So that give a good 6 month to amd to take a lead over northwood/Press.
The worse 31 stage pipeligne for XEON will kill it maybe that why the rumor say that XEON will have larger L1 cache.A64 might not be a scotty killer but actual opteron will take a giant lead over XEON.
They have take the long way like they allwayse do for more performance the short way more cache improve some stuff more clock speed would have put amd in big probleme.Now if they lose the 32 bit war itanium will fall and there market share.
I dont like french test
So that give a good 6 month to amd to take a lead over northwood/Press.
The worse 31 stage pipeligne for XEON will kill it maybe that why the rumor say that XEON will have larger L1 cache.A64 might not be a scotty killer but actual opteron will take a giant lead over XEON.
They have take the long way like they allwayse do for more performance the short way more cache improve some stuff more clock speed would have put amd in big probleme.Now if they lose the 32 bit war itanium will fall and there market share.
I dont like french test
I've read your arguements back and forth, amd vs intel, for 6 pages and it dosen't really matter. Both amd and intel's new cpu's are babies, towards the end of this year we will see what difference the choice of design makes. Personally I buy amd as they have been somewhat better in price/performance. But I suspect that by christmas we will see amd trailing intel in the benchmarks again. I don't say this because I think intel has a superior design, but because they have a more aggresive road map then amd. If amd truly wants to be king of the hill like they did with the orginal athlon then they need to increase cpu speed on a faster rate then what they are currently doing. Intel has made a very smart move with the prescott, it performes very much the same as the northword so to a consumer why not buy it for the future sse3 instrutions. And as cpu speed and frontside bus speed increases we are going to see "a coming of age" for the prescott as we did the P4 from when it was first introduced.
> you must be used to efficiency numbers for thermal
>devices, which are all inherently efficiency-limited
Really ? well I'm no physicist but doh.. I'd think these are 100% efficient, since all the electrical energy gets converted in heat.. or am I missing something ?
>The 75% figure seems more credible
Actually, I looked up some specs from Zalman and other PSU manufacturers, and they state around 75% under full load. Since you wouldnt (want to) run those at full load, effeciency ought be less, but I don't know by how much either.
>which is equivalent to changing two, maybe three common
>tungsten light bulbs to power-saving lights.
I'll be caught dead before I use those power saving lights
I'd rather just switch them off, I hate that cold unnatural light. But I guess that is a bit OT.
>It's simply not an issue, if you consider showers (3KW)
I don't shower 24/7, I do run my computer 24/7, and its under full load nearly half of that (crunching DivX>MP2 or)
>or
>electrical irons (1+KW)
I do use that maybe 1 hour a week- at most
>or refrigerators/freezers
True, and you could add a LOT of things to that list, but I still do look at power consumption when buying a freezer or washing machine, while my computer as a whole may actually consume about as much.
Anyway, do not think I am claiming this to be the most relevant purchase criterium, by a long shot, but I still like to have a computer that is as silent as possible, and that doesnt burn more electricity than necessary. The cost might not be an issue, but Cool&Quiet is actually a strong argument for me to consider a A64. But sure, being faster, cheaper and more future proof is also part of the equation 😉
= The views stated herein are my personal views, and not necessarily the views of my wife. =
>devices, which are all inherently efficiency-limited
Really ? well I'm no physicist but doh.. I'd think these are 100% efficient, since all the electrical energy gets converted in heat.. or am I missing something ?
>The 75% figure seems more credible
Actually, I looked up some specs from Zalman and other PSU manufacturers, and they state around 75% under full load. Since you wouldnt (want to) run those at full load, effeciency ought be less, but I don't know by how much either.
>which is equivalent to changing two, maybe three common
>tungsten light bulbs to power-saving lights.
I'll be caught dead before I use those power saving lights
I'd rather just switch them off, I hate that cold unnatural light. But I guess that is a bit OT.>It's simply not an issue, if you consider showers (3KW)
I don't shower 24/7, I do run my computer 24/7, and its under full load nearly half of that (crunching DivX>MP2 or)
>or
>electrical irons (1+KW)
I do use that maybe 1 hour a week- at most
>or refrigerators/freezers
True, and you could add a LOT of things to that list, but I still do look at power consumption when buying a freezer or washing machine, while my computer as a whole may actually consume about as much.
Anyway, do not think I am claiming this to be the most relevant purchase criterium, by a long shot, but I still like to have a computer that is as silent as possible, and that doesnt burn more electricity than necessary. The cost might not be an issue, but Cool&Quiet is actually a strong argument for me to consider a A64. But sure, being faster, cheaper and more future proof is also part of the equation 😉
= The views stated herein are my personal views, and not necessarily the views of my wife. =
>Only the unregistered/registered part will make a
>difference there now, won't it?
For the FX, probably not, though I did read about a new stepping being launched with s939 that would increase SSE2 and/or memory performance as well. Probably noting major on its own, but since we are talking about cpu's that perform within a few percent of each other, every little bit helps.. We'll see soon enough though.
>So the benefit must lie in the very few percentage points
>from unbuffered RAM. I gave you my guestimate, time will prove me right or wrong.
I dug up some numbers for you. Have a look here:
<A HREF="http://www.x86-secret.com/articles/cpu/a64_3000/a64_3000-3.htm/" target="_new">http://www.x86-secret.com/articles/cpu/a64_3000/a64_3000-3.htm/</A>
THe biggest difference is going from ECC to non ECC, but if you compare registered+ecc to non registered+non ecc, differences can get a as big as 10% in both latency and throughput. THey only test one game bench (unreal) and with no detailed info and surprisingly low framerates, its a bit guessing, but that 3% delta in FPS actually correspond roughly with a speedgrade difference when I look at Tom's UT benches.
>P4 chipsets in general are old, but LGA775 and DDR2
>platforms with PCI-E will be brand-spanking new
New doesnt mean faster; just look at prescott. Like I said, neither PCI-E nor DDRII will give a performance boost IMHO.
>A64 will scale well, but Prescott's scaling is still very
>much a mistery. And even so, some parts will be "Manually
>scaled", like FSB and memory, within the next few months.
Correct, but memory will also be "scaled manually" for the A64. Prescotts fsb increase will probably help it not loosing too much from A64 as clockspeeds increase though, I agree with that. Especially as its FSB will be increased more than its clock.
>Intel sometimes surprises us
Oh yeah. Like with willamette, rdram, prescott Seriously though, the only real rabbit I can see them pull out their hat is fixing whatever they messed up with prescott in the first place. And I'm guessing Prescott was supposed to have dynamic multithreading, which allows multithreading single threaded apps, and would make a lot of sense on such a deep pipelined cpu, BUT I'm ALSO guessing that is complex enough not to be able to fix it with just a stepping. Tejas might be interesting, but I wouldnt hold my breath for Prescott. Given the extreme low ammount of hype even intel is producing for Prescott or its new instruction, I think they gave up on it and now concentrate on a 64 bit DHT Tejas instead. Seems like DHT might be nice btw, and have a serious potential.
>Read: AMD still has a shrink to do, not only to take
>advantage of. It still has to be done,
Yes, which was my point; they may not even require it to keep at least parity with intel this year. AMD also seems very confident on and happy with their 90nm process, I'm not too worried.
> By then, Intel will have had time to update the prescott
>core for further scaling and upgrade their manufacturing
>techs.
I think there is nothing wrong with intel's 90nm process. Dothan will give us certainty, but I think everything that sucks about Prescott, is because of prescott, not the 90nm process.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
>difference there now, won't it?
For the FX, probably not, though I did read about a new stepping being launched with s939 that would increase SSE2 and/or memory performance as well. Probably noting major on its own, but since we are talking about cpu's that perform within a few percent of each other, every little bit helps.. We'll see soon enough though.
>So the benefit must lie in the very few percentage points
>from unbuffered RAM. I gave you my guestimate, time will prove me right or wrong.
I dug up some numbers for you. Have a look here:
<A HREF="http://www.x86-secret.com/articles/cpu/a64_3000/a64_3000-3.htm/" target="_new">http://www.x86-secret.com/articles/cpu/a64_3000/a64_3000-3.htm/</A>
THe biggest difference is going from ECC to non ECC, but if you compare registered+ecc to non registered+non ecc, differences can get a as big as 10% in both latency and throughput. THey only test one game bench (unreal) and with no detailed info and surprisingly low framerates, its a bit guessing, but that 3% delta in FPS actually correspond roughly with a speedgrade difference when I look at Tom's UT benches.
>P4 chipsets in general are old, but LGA775 and DDR2
>platforms with PCI-E will be brand-spanking new
New doesnt mean faster; just look at prescott. Like I said, neither PCI-E nor DDRII will give a performance boost IMHO.
>A64 will scale well, but Prescott's scaling is still very
>much a mistery. And even so, some parts will be "Manually
>scaled", like FSB and memory, within the next few months.
Correct, but memory will also be "scaled manually" for the A64. Prescotts fsb increase will probably help it not loosing too much from A64 as clockspeeds increase though, I agree with that. Especially as its FSB will be increased more than its clock.
>Intel sometimes surprises us
Oh yeah. Like with willamette, rdram, prescott
Seriously though, the only real rabbit I can see them pull out their hat is fixing whatever they messed up with prescott in the first place. And I'm guessing Prescott was supposed to have dynamic multithreading, which allows multithreading single threaded apps, and would make a lot of sense on such a deep pipelined cpu, BUT I'm ALSO guessing that is complex enough not to be able to fix it with just a stepping. Tejas might be interesting, but I wouldnt hold my breath for Prescott. Given the extreme low ammount of hype even intel is producing for Prescott or its new instruction, I think they gave up on it and now concentrate on a 64 bit DHT Tejas instead. Seems like DHT might be nice btw, and have a serious potential.>Read: AMD still has a shrink to do, not only to take
>advantage of. It still has to be done,
Yes, which was my point; they may not even require it to keep at least parity with intel this year. AMD also seems very confident on and happy with their 90nm process, I'm not too worried.
> By then, Intel will have had time to update the prescott
>core for further scaling and upgrade their manufacturing
>techs.
I think there is nothing wrong with intel's 90nm process. Dothan will give us certainty, but I think everything that sucks about Prescott, is because of prescott, not the 90nm process.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
Oh, and Fisto by the way; I hope you realize your assumptions/hopes generally require intel to exceed what their roadmaps predict (like "fixing" prescott, and other magical enhancements) and require AMD to falter (like missing their 90nm process, not being able to increase clockspeed of K8, etc). Anything is possible, but the combination of both just doesnt seem very likely (and problably wouldnt even give intel the upper hand either).
I see this all the time, cause intel is bigger and meaner, people just won't believe the facts even when they are looking at them. Seen it with 820/840 ("RDRAM will give a huge boost with willamette and prices will drop to SDRAM levels").
seen it with willamette ("SSE2 will magically turn it into a screamer"), with prescott, with Itanium.. I find that attitude intruiging. Throughout history, not one company has maintained a technological lead or dominant market position for the duration of its history. intel is not immune to this, neither is even mircosoft. The only question is: when ? Well how about *now* for intel loosing technological leadership when it comes to x86 ? they are outclassed in the high end (x86 server/ws), got a grim perspective in the mainstream (desktop) market, they can only claim leadership in the mobile market -for now; I can't wait to see how Transmeta's Efficion performs If this lasts for too long, it will sooner or later have serious consequences for its market share, no matter how competent intel's PR is, or how loyal its fanboys (not necessarely including you here).
= The views stated herein are my personal views, and not necessarily the views of my wife. =
I see this all the time, cause intel is bigger and meaner, people just won't believe the facts even when they are looking at them. Seen it with 820/840 ("RDRAM will give a huge boost with willamette and prices will drop to SDRAM levels").
seen it with willamette ("SSE2 will magically turn it into a screamer"), with prescott, with Itanium.. I find that attitude intruiging. Throughout history, not one company has maintained a technological lead or dominant market position for the duration of its history. intel is not immune to this, neither is even mircosoft. The only question is: when ? Well how about *now* for intel loosing technological leadership when it comes to x86 ? they are outclassed in the high end (x86 server/ws), got a grim perspective in the mainstream (desktop) market, they can only claim leadership in the mobile market -for now; I can't wait to see how Transmeta's Efficion performs
If this lasts for too long, it will sooner or later have serious consequences for its market share, no matter how competent intel's PR is, or how loyal its fanboys (not necessarely including you here).= The views stated herein are my personal views, and not necessarily the views of my wife. =
- Feb 10, 2003
- 2,444
- 0
- 19,780
I just woke up, and realized some things I wrote actually didn't make much sense or were too biased. I think I got tired from reading all reviews and wasn't thinking straight. Sorry.
But, hey, a new day and a clear view... it's just hardware.
BTW, sorry about that "bob" thing too, it's just that I think someone called you bob around here, and for some reason, I started using that too. Won't happen again.
And now, with my refreshed view on things... Prescott is a funny processor, isn't it? :lol:
<font color=red><b>M</b></font color=red>ephistopheles<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 02/02/04 10:17 AM.</EM></FONT></P>
But, hey, a new day and a clear view... it's just hardware.
BTW, sorry about that "bob" thing too, it's just that I think someone called you bob around here, and for some reason, I started using that too. Won't happen again.
And now, with my refreshed view on things... Prescott is a funny processor, isn't it? :lol:
<font color=red><b>M</b></font color=red>ephistopheles<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 02/02/04 10:17 AM.</EM></FONT></P>>BTW, sorry about that "bob" thing too,
Hey, its my name.. I can understand "bbaeyens" isnt quite easy to type, and heck, I whish I had chosen a different nick all these years ago, but I can't change it, tough luck. feel free to abreviate it to "bb" though.
>And now, with my refreshed view on things... Prescott is a
>funny processor, isn't it?
Why did you edit that part ? I prefered what you wrote earlier: it sucks..
= The views stated herein are my personal views, and not necessarily the views of my wife. =
Hey, its my name.. I can understand "bbaeyens" isnt quite easy to type, and heck, I whish I had chosen a different nick all these years ago, but I can't change it, tough luck. feel free to abreviate it to "bb" though.
>And now, with my refreshed view on things... Prescott is a
>funny processor, isn't it?
Why did you edit that part ? I prefered what you wrote earlier: it sucks..
= The views stated herein are my personal views, and not necessarily the views of my wife. =
This is strange. The new p4 is hotter, and no faster than the northwood. It is intended as a mid range product (according to the review). So why would the average joe buy it? Its sad but the fact is when choosing a computer, the average guy sees 3.2 is higher than an amd 2.2, so buys the 3.2 p4. So, knowing this, why launch a proccessor which is slower than this 3.2? He's not going to buy it, it has a lower number next to the 'ghz' :|
I guess intel launched this so they could make cheaper 2.8-3.2 ghz cpus (due to 90nm) and compete with amd in that respect? I think if this is the case, it is unwinable. Those who buy amd (enthusiasts, students, etc) will continue to do so, and those who jump at the high clock speeds will continue to do so. I'm no AMD fanboy, but I do like to see the market leader launching a dud product - if that is the case - as amd can close up the clock speed war, which they need to in marketing terms, not for performance - as above, average man wont buy a 64 athlon if its only 2.2 compared to the p4 @3.2. IMO they need to match numbers with intel, catch up, at least to some degree.
ps this is just my opinion, if you plan on ripping it to pieces (as is your right 😛) then be polite
Cheers,
Chris
XP2000, 256ddr 2100ram, GF4 MX440, XP Pro
I guess intel launched this so they could make cheaper 2.8-3.2 ghz cpus (due to 90nm) and compete with amd in that respect? I think if this is the case, it is unwinable. Those who buy amd (enthusiasts, students, etc) will continue to do so, and those who jump at the high clock speeds will continue to do so. I'm no AMD fanboy, but I do like to see the market leader launching a dud product - if that is the case - as amd can close up the clock speed war, which they need to in marketing terms, not for performance - as above, average man wont buy a 64 athlon if its only 2.2 compared to the p4 @3.2. IMO they need to match numbers with intel, catch up, at least to some degree.
ps this is just my opinion, if you plan on ripping it to pieces (as is your right 😛) then be polite
Cheers,
Chris
XP2000, 256ddr 2100ram, GF4 MX440, XP Pro
- Feb 10, 2003
- 2,444
- 0
- 19,780
You know, I wish I had chosen a different nick too... Mephistopheles is just way too complicated. It's actually the name of the Devil in Goethe's Faust - I once played that role! It was great fun. One of the greatest things I ever did. And Goethe writes german in such a way that, once you've learned it by heart, you can really appreciate the sound and impact of every word in his text. (Besides, being the Devil is always fun. We had a lady god... she wasn't any good. I'd much rather be <font color=red>the devil</font color=red> instead.)
The "normal" - if there is such a thing - abbreviation for Mephistopheles is Mephisto. But I don't spread that around so much, because it's fun to see what other people make up! I've been called Meph, Mephy, Fisto...
As for prescott sucking... Well, it does indeed. :lol:
But I edited that because I still think that, as stupid as it would be to buy a prescott now, Scotty still will scale well enough to make a worthy processor. It won't matter if it's a whole speed grade below Northwood later, when it's scaled to 4Ghz and all trimmed up.
Right now, though, it stinks.
<font color=red><b>M</b></font color=red>ephistopheles
The "normal" - if there is such a thing - abbreviation for Mephistopheles is Mephisto. But I don't spread that around so much, because it's fun to see what other people make up! I've been called Meph, Mephy, Fisto...
As for prescott sucking... Well, it does indeed. :lol:
But I edited that because I still think that, as stupid as it would be to buy a prescott now, Scotty still will scale well enough to make a worthy processor. It won't matter if it's a whole speed grade below Northwood later, when it's scaled to 4Ghz and all trimmed up.
Right now, though, it stinks.
<font color=red><b>M</b></font color=red>ephistopheleshmm.. how to rip this appart politely ... ?
😀
I disagree though.. your average Joe will not see the clockspeed of the AMD's, they will see the PR number. Hell, I know I have to think twice what the clock speed is of a FX53 or 3400+. If AMD would ever catch intel in clockspeed (very, very, utterly unlikely), intel is history in the performance market. A 3.2 GHz Athlon 64 may score in the 5000+ range. not gonna happen.
Secondly, if marketed well, Joe will see a big number, and 64 bit.. twice as much as 32 bit, and he may well remember the 64 bit console hype, or the "32 bit" windows95 launch which taught him: more bits is better.
>The new p4 is hotter, and no faster than the northwood. It
>is intended as a mid range product (according to the
>review). So why would the average joe buy it?
Joe will buy whatever Dell/HP/Packard Bell sells him. he won't know or care if it is a P4E or C. So that, I think is not intel's problem. I think Packard Bell or Dell may not be happy to have to pay for more expensive PSU's, motherboards and cooling solutions for a mid range product (once Prescott goes mainstream). they would make more profit using other cpu's. That would be intels problem. Even more so, if prescott based Celerons turn out to suffer the same illness.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
😀
I disagree though.. your average Joe will not see the clockspeed of the AMD's, they will see the PR number. Hell, I know I have to think twice what the clock speed is of a FX53 or 3400+. If AMD would ever catch intel in clockspeed (very, very, utterly unlikely), intel is history in the performance market. A 3.2 GHz Athlon 64 may score in the 5000+ range. not gonna happen.
Secondly, if marketed well, Joe will see a big number, and 64 bit.. twice as much as 32 bit, and he may well remember the 64 bit console hype, or the "32 bit" windows95 launch which taught him: more bits is better.
>The new p4 is hotter, and no faster than the northwood. It
>is intended as a mid range product (according to the
>review). So why would the average joe buy it?
Joe will buy whatever Dell/HP/Packard Bell sells him. he won't know or care if it is a P4E or C. So that, I think is not intel's problem. I think Packard Bell or Dell may not be happy to have to pay for more expensive PSU's, motherboards and cooling solutions for a mid range product (once Prescott goes mainstream). they would make more profit using other cpu's. That would be intels problem. Even more so, if prescott based Celerons turn out to suffer the same illness.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
- Feb 10, 2003
- 2,444
- 0
- 19,780
That was a funny one! :lol:if you plan on ripping it to pieces (as is your right 😛) then be polite
But unfortunately, I think bbaeyens has a point.
Have PR ratings really hurt AMD's Athlon XP?... I don't really think so.
In essence, the average joe will still think that Intel and AMD are about equal right now, because of numbers.
I was just thinking... What if Scotty is Tejas' Willamette? As in northwood being the good Willamette sibling, and Tejas will improve on Scotty?
I'm not excusing Prescott's sub-par performance at all here...You guys be polite with me too. 😱
<font color=red><b>M</b></font color=red>ephistopheles<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 02/02/04 11:05 AM.</EM></FONT></P>>I was just thinking... What if Scotty is Tejas' Willamette?
Quite possible; I have *no* idea what to expect from Tejas, and I think precious few people do.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
Quite possible; I have *no* idea what to expect from Tejas, and I think precious few people do.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
BigMac
Splendid
- Nov 25, 2003
- 5,627
- 0
- 25,780
power consumption is part of the costs. It's like saying "I do not care what the mileage on my car is, as long as operational costs are low", whereas the fuel consumption of your car is an important factor in determining those operational costs.
Bottomline, if the power consumption goes through the roof, you will have a hard time to reduce costs elsewhere to come up with an overal good performance to cost ratio.
Granted, with fuel these effects vary whether you live in the US or in Europe as fuel prices vary greatly (and this is reflected in fuel consumption statistics of new cars being sold in US and Europe).
Price variability probably also holds true for energy prices in general (although I am not in the know on that one) but lucky for us, PC's have already standardized globally. So indeed either power consumption is not such a hot issue (pun intended) as it is with cars, or the US is dominating the market, making it less of an issue at this time (my guess is the latter).
I will refrain from making an obvious remark concerning laptops.
Ps: woops, thats what you get when pressing return after starting a reply more than 6 hours ago. O well, hopefully this post contains something worthwhile, if not discard it...
PPs: I've used "Bob" on a number of occasions as it's the guys name according to his profile. Is it against netiquette to use someone's published first name instead of his forum nickname? If so, please accept my apologies (but I am not aware of any such "rule").
BigMac
<A HREF="http://www.p3int.com/product_center_NWO_The_Story.asp" target="_new">New World Order</A><P ID="edit"><FONT SIZE=-1><EM>Edited by BigMac on 02/02/04 02:48 PM.</EM></FONT></P>
- Feb 10, 2003
- 2,444
- 0
- 19,780
Well, true, I exaggerated a bit there.
But my point is that prescott being 90/100/110W won't do any real and true harm in your typical power bill in terms of percentage increase, unless you run 24/7. And I'd abdicate from keeping my computer on at night before abdicating on buying the eventual 125W processor, if it came with appropriate performance.
<font color=red><b>M</b></font color=red>ephistopheles<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 02/02/04 01:21 PM.</EM></FONT></P>
But my point is that prescott being 90/100/110W won't do any real and true harm in your typical power bill in terms of percentage increase, unless you run 24/7. And I'd abdicate from keeping my computer on at night before abdicating on buying the eventual 125W processor, if it came with appropriate performance.
<font color=red><b>M</b></font color=red>ephistopheles<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 02/02/04 01:21 PM.</EM></FONT></P>>But my point is that prescott being 75/100/125W won't do
>any real and true harm in your typical power bill
Paying $1000 for a cpu won't do any real and true harm in my typical yearly expenses/revenues either :/
= The views stated herein are my personal views, and not necessarily the views of my wife. =
>any real and true harm in your typical power bill
Paying $1000 for a cpu won't do any real and true harm in my typical yearly expenses/revenues either :/
= The views stated herein are my personal views, and not necessarily the views of my wife. =
after reading tom's review and anadtech review, i fount anad's review more into details of the processor, more into technical details, anyway, some of intels fan are true for the prescott, and amd's fan are more logicly talking for the ups and downs of the prescott.
Intel did tell that the prescott is not the top processor for this moment, its true, but it would take the lead of intel processors in the future.
Here i found intel in a good step for future, but not good enough to go over AMD, why?
first thing the northwood had 20 stages, and the prescott have 31 stages, that means that prescott will perform a lot slower than noethwood at the same clock, if they have the same specification (like caches and.......).
but here so to prescott can equal northwood performance at the same clock speed, they increased the l1 and l2 cache size, they introduced in it the sse3, they upgraded a little the HT technology, and they produced it in 90nm.
So this updates succeeded to equal the northwood. But prescott will shine when they will be clocked to higher clock frequencies, cos it can be clocked a lot higher, that is what they do it, and increased to stages, and switched to 90nm.
Anyway the architechture of this prossecor is to reach higher clock speeds than nothwood, so it will beet it with performance.
But there is something, the prescott is toooooo hot, it will need powerfull cooling, and i think that heat will limit its clock frequency later.So thats a big down for tehm, as they need more colling, thus more wattage from the power supply, and a lot more noise. (audio users will be disapointed with it). and the sse3 is not a big improvement as did the sse2 to the sse1. And it stills a 32 bit cpu, its not a big problem, but companies like intel that love marqueting, they will loose in this one this time.
Better talking prescott with its new platform, the new chipsets will be better for it, as it is designed for them, we will see for sure a performance increase (plus increase in fsb, and bla bla bla).
As for AMD they are moving more smartly, as for my point of view. why??????????/
Amd not like intel always tries to get something from the precessor withought raising clock frequencies. see the difference between an fx51 and a athlon xp 3200+ (both clocked at 2.4ghz). this is what im talking about, and they are working alot with power consemption and cool&quiet for quiet processing, thats a verry good point, cos it attracts silent pc users more than intels processor for less money.
Amd have also a lot of rooms to grow with it, and espacially it scales a lot better with the increase of clock speed, they will introduce the 90nm process, and go up in frequencies, while updating to ddr2 or ddr533(ddr1), and pci express, and perhaps an sse3 (no body nows).
Thats my point of view, I will stuck to AMD now, its safer for future then scotty.
<P ID="edit"><FONT SIZE=-1><EM>Edited by parhelia on 02/02/04 04:07 PM.</EM></FONT></P>
Intel did tell that the prescott is not the top processor for this moment, its true, but it would take the lead of intel processors in the future.
Here i found intel in a good step for future, but not good enough to go over AMD, why?
first thing the northwood had 20 stages, and the prescott have 31 stages, that means that prescott will perform a lot slower than noethwood at the same clock, if they have the same specification (like caches and.......).
but here so to prescott can equal northwood performance at the same clock speed, they increased the l1 and l2 cache size, they introduced in it the sse3, they upgraded a little the HT technology, and they produced it in 90nm.
So this updates succeeded to equal the northwood. But prescott will shine when they will be clocked to higher clock frequencies, cos it can be clocked a lot higher, that is what they do it, and increased to stages, and switched to 90nm.
Anyway the architechture of this prossecor is to reach higher clock speeds than nothwood, so it will beet it with performance.
But there is something, the prescott is toooooo hot, it will need powerfull cooling, and i think that heat will limit its clock frequency later.So thats a big down for tehm, as they need more colling, thus more wattage from the power supply, and a lot more noise. (audio users will be disapointed with it). and the sse3 is not a big improvement as did the sse2 to the sse1. And it stills a 32 bit cpu, its not a big problem, but companies like intel that love marqueting, they will loose in this one this time.
Better talking prescott with its new platform, the new chipsets will be better for it, as it is designed for them, we will see for sure a performance increase (plus increase in fsb, and bla bla bla).
As for AMD they are moving more smartly, as for my point of view. why??????????/
Amd not like intel always tries to get something from the precessor withought raising clock frequencies. see the difference between an fx51 and a athlon xp 3200+ (both clocked at 2.4ghz). this is what im talking about, and they are working alot with power consemption and cool&quiet for quiet processing, thats a verry good point, cos it attracts silent pc users more than intels processor for less money.
Amd have also a lot of rooms to grow with it, and espacially it scales a lot better with the increase of clock speed, they will introduce the 90nm process, and go up in frequencies, while updating to ddr2 or ddr533(ddr1), and pci express, and perhaps an sse3 (no body nows).
Thats my point of view, I will stuck to AMD now, its safer for future then scotty.
<P ID="edit"><FONT SIZE=-1><EM>Edited by parhelia on 02/02/04 04:07 PM.</EM></FONT></P>
TRENDING THREADS
-
-
Question Help games are crashing and my hair is being pulled out- Started by Joe Gallo
- Replies: 17
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
-
-
Facebook
Twitter
Instagram