All you have to do is read the mags and the online reviews and it shows the p4 isnt all that.its that plain and simple.Its a good chip when used with the new sse2.Also dont think amd will sit on its ass.it hasnt gotten this far by just sitting here.
Insane3d Very cool! Shows alot of soul. I like that, let him be happy just let it go. Very soul! By the way I run a 733 celeron love it works for me, but run what you have. Be cool! Nos440 I'm happy for you sounds like you are into it. You have a great machine. We all have great machines and our next one will be greater for what we put into it. Now I'm off to do yoga. (LOL)
Flanders out!
Flanders out!
Insane3d Very cool! Shows alot of soul. I like that, let him be happy just let it go. Very soul! By the way I run a 733 celeron love it works for me, but run what you have. Be cool! Nos440 I'm happy for you sounds like you are into it. You have a great machine. We all have great machines and our next one will be greater for what we put into it. Now I'm off to do yoga. (LOL)
Flanders out!
Flanders out!
<< I have a IDE Raid card , Network Card , Hollywood Plus DVD card , STB TV/FM card / SB Live and of coarse a geforce 2 Ultra (no special case or power Supply needed). I've run it for 2 days solid without a hitch its taken everything I could throw at it and its clocked to 1708 mhz with a bony a$$ $9 dollar cooler LOL !!! It runs windows like a mad man and Games are a blur its silky smooth at everything. >>
That's funny... I get the same results from my AMD Duron rig. You spent how much for that P4? Wish I had that kind of money to waste.
How's this for a price comparison....
I'm speaking on behalf of all the Anandtechers who live in Australia...
Average P4 1.5GHz Price: $2300AUD
Average TBird 1.2GHz Price: $580AUD
Hmm... for a processor to beat or match another in many many many programs, and come in at 1/4 the price is a pretty marvellous job!
My PII400, on the other hand, is a beast
I'm speaking on behalf of all the Anandtechers who live in Australia...
Average P4 1.5GHz Price: $2300AUD
Average TBird 1.2GHz Price: $580AUD
Hmm... for a processor to beat or match another in many many many programs, and come in at 1/4 the price is a pretty marvellous job!
My PII400, on the other hand, is a beast
funny thing about computers is that there always something that comes out faster. Couple of weeks something will come out faster then intel and then faster then that. Then this conversation will be useless.
Btw, there's one argument not mentioned so far which definitely speaks for a T-Bird: raw FPU performance. Even with all the SSE, 3Dnow! optimizations, there are still applications where you need a strong FPU, scientific (=numerical) calculations for example. And the Athlon FPU is undoubtedly much better than the PIII-FPU and the P4-FPU. So, it was no question for me...Athlon was my choice, because I have to do such calculations quite often.
BTW, unless you buy that Asus P4 motherboard that comes with a baseplate, you do need a special P4 complient case, otherwise you can't mount the Heatsink the required way.
Just check the any of the bigger reveiws on the P4 (Tom's, Anand, Sharky, etc)
Just check the any of the bigger reveiws on the P4 (Tom's, Anand, Sharky, etc)
Mem
Lifer
- Apr 23, 2000
- 21,476
- 13
- 81
NOS440,Just to let you know there are very happy T-bird owners out there,Yes I`m one of them,I`m also a gamer & do RC5 etc,my system is rock solid & alot cheaper then what you payed,true yours is faster,but everybody knows what`s fast today is slow Tomorrow,& as I said I`m very happy with the stability which is & always will be my first priorty..
MadRat
Lifer
- Oct 14, 1999
- 11,999
- 307
- 126
When you compare the P4 to the Thunderbird try opening up multiple instances of Microsoft Office products, like Access and Excel. Run a query in access against two 100mb databases across your network while simultaneously running Word or Excel and then try to do work in the latter. Let NOS440 explain the results here.
The P4 is a single task wonder. The P!!! was also very fast at single tasks. The problem is most people in the IT world run ten things at once. AMD figured that out and tailored their processors since the K6 to do heavy multitasking. Benchmarks just do not show the potential of heavy multitasking.
The P4 is a single task wonder. The P!!! was also very fast at single tasks. The problem is most people in the IT world run ten things at once. AMD figured that out and tailored their processors since the K6 to do heavy multitasking. Benchmarks just do not show the potential of heavy multitasking.
<< I'm on a roll and you guy's are pissed I love it >>
Did somebody say "troll" .. uhm no? guess I was hearing those stange voices again
<< The P4 is a single task wonder. The P!!! was also very fast at single tasks. The problem is most people in the IT world run ten things at once. AMD figured that out and tailored their processors since the K6 to do heavy multitasking. Benchmarks just do not show the potential of heavy multitasking. >>
That is a very interesting point... it does indeed comply with my experinces with various K6/K7 vs. P6 based system
Well, as a processor, IMHO P4 is pathetic. He lives only on huge memory bandwith and SSE2 optimation. Take those two things away from it and you've got chip that performs like VIA Cyrix III. P4's brute FPU and integer power is lower than you find from P6 core and way below athlon's. Most of my computing power goes into RC-5 and I'm not gonna downgrade my main box from Duron700 to P4/1,5Ghz any time soon
MadRat
Lifer
- Oct 14, 1999
- 11,999
- 307
- 126
Intel staggered the four 100mHz buses like it staggered the L2 caches on the P!!!. This means that its a 100mHz bus each clock cycle, with only one available per cycle. It has to query all four buses to find the data it wants.
They could have split the serial output of each memory module to quntify the data available each cycle. Instead they took the easy way out, choosing the simpler design. They misrepresent the memory bandwidth of the P4 design by implying all four buses operate in unison. Trouble is there will never be the 5gb/sec of bandwidth available like they claim. Rather, they will never have more than the potential of ONE memory module available at any one time.
It will take incredibly high clock speeds in order to mask the memory latency of this design. Don't get me wrong, more "potential bandwidth" is available on a qualitative scale, just not on each cycle. It takes four cycles to work through the four memory buses. After 1000 cycles, if you factor in latency per each chipset design (QDR, DDR, etc.), the P4 bus will have been able to push more data through. If it takes four cycles to complete the search then where is its advantage?
I think the P4's reliance on system memory (factoring in its L1 design and lack of L3 cache) will always be its bane. Too many times the processor is going to be looking into the system memory for data. The next generation of P4 will undoubtedly snap this barrier, but for now Intel is stuck with a lame duck.
They could have split the serial output of each memory module to quntify the data available each cycle. Instead they took the easy way out, choosing the simpler design. They misrepresent the memory bandwidth of the P4 design by implying all four buses operate in unison. Trouble is there will never be the 5gb/sec of bandwidth available like they claim. Rather, they will never have more than the potential of ONE memory module available at any one time.
It will take incredibly high clock speeds in order to mask the memory latency of this design. Don't get me wrong, more "potential bandwidth" is available on a qualitative scale, just not on each cycle. It takes four cycles to work through the four memory buses. After 1000 cycles, if you factor in latency per each chipset design (QDR, DDR, etc.), the P4 bus will have been able to push more data through. If it takes four cycles to complete the search then where is its advantage?
I think the P4's reliance on system memory (factoring in its L1 design and lack of L3 cache) will always be its bane. Too many times the processor is going to be looking into the system memory for data. The next generation of P4 will undoubtedly snap this barrier, but for now Intel is stuck with a lame duck.
MadRat...although your theory here sounds really good but in every Memory intensive benchmark I have seen it doesn't hold true the P-4 System out performs the DDR thunderbird by a sizeable amount can you explain this to me ?? (I bolded the System word so people realize this is about the syatem as a whole). Also I have multi tasked with this system without any problems I have run a 3 gig data transfer from PCs across my network burned a CD with my Plextor IDE 12X and installed a program and also was on the net with IE with Multiple pages and also had AOL Instant messenger running without a problem explain somemore.
Priit...your right but brute force is a real stone age way of doing things isn't it. The modern way to make a system work is a trade off here for something there the P-4 is a well balanced system and is now doing well in most Apps that performance is needed and there aren't even optimised with SSE-2 yet !!!!!
Zephyr...that was a good one I have hand it to ya there !!!! but some people don't understand are twisted relationship and history !!
Mem... I'm sure you are happy and have a right to be the T-Bird is a nice system But it has met its Better.. Also for your imformation this uprade after the sale of my existing motherboard , CPU and Ram cost me a whopping 70 Bucks. I realize that most will not get the deal I got But that is not what I'm talking about here why does everyone so hung up on price is this all that you people think about ???
DABANSHEE...wrong MSI just released a board that does it the same way as Asus and Abit will be next. besides the Asus is a awesome board.
Compellor... read reply to Men above....LOL I'm happy as a Jbird with the upgrade.
Hanky I agree But there are no programs that will show us yet what the SSE-2 will have in store for the P-4. Well except the one that Toms Hardware did and if this is any indication the P-4 is going to rock the T-bird world there too !!! But there is no fact to this so lets not slam me for it guys.
PieDerro the price in AUD really means nothing to me. I'm sorry that it cost that much where you are and yes you should buy a T-Bird at that price difference.
DarkMajiq......don't bother pittying me I'm in heaven here on my P-4 ......
Insane3D I'm glad your taking this in a better light although still throwing insults. This is all fun to me and yes my system is better then yours
Dexion..... say's you *&%$E()*&^_{_
Man this was a busy night and morning for you all I was happily sleeping keep it all coming .......
Priit...your right but brute force is a real stone age way of doing things isn't it. The modern way to make a system work is a trade off here for something there the P-4 is a well balanced system and is now doing well in most Apps that performance is needed and there aren't even optimised with SSE-2 yet !!!!!
Zephyr...that was a good one I have hand it to ya there !!!!
but some people don't understand are twisted relationship and history !!Mem... I'm sure you are happy and have a right to be the T-Bird is a nice system But it has met its Better.. Also for your imformation this uprade after the sale of my existing motherboard , CPU and Ram cost me a whopping 70 Bucks. I realize that most will not get the deal I got But that is not what I'm talking about here why does everyone so hung up on price is this all that you people think about ???
DABANSHEE...wrong MSI just released a board that does it the same way as Asus and Abit will be next. besides the Asus is a awesome board.
Compellor... read reply to Men above....LOL
I'm happy as a Jbird with the upgrade.Hanky I agree But there are no programs that will show us yet what the SSE-2 will have in store for the P-4. Well except the one that Toms Hardware did and if this is any indication the P-4 is going to rock the T-bird world there too !!! But there is no fact to this so lets not slam me for it guys.
PieDerro the price in AUD really means nothing to me. I'm sorry that it cost that much where you are and yes you should buy a T-Bird at that price difference.
DarkMajiq......don't bother pittying me I'm in heaven here on my P-4 ......
Insane3D I'm glad your taking this in a better light although still throwing insults. This is all fun to me and yes my system is better then yours
Dexion..... say's you *&%$E()*&^_{_
Man this was a busy night and morning for you all I was happily sleeping keep it all coming .......
<< MadRat...although your theory here sounds really good but in every Memory intensive benchmark I have seen it doesn't hold true the P-4 System out performs the DDR thunderbird by a sizeable amount can you explain this to me ?? (I bolded the System word so people realize this is about the syatem as a whole). Also I have multi tasked with this system without any problems I have run a 3 gig data transfer from PCs across my network burned a CD with my Plextor IDE 12X and installed a program and also was on the net with IE with Multiple pages and also had AOL Instant messenger running without a problem explain somemore. >>
I don't see such a big deal, any processor over 700Mhz can easily do this as long as there isn't any software problems ie: microsoft. Infact, what your experiencing isn't relied on memory bandwidth at all. Its irrelevant. Infact, I doubt you can distinguish each platform (P3,P4,Athlon,Duron) with desktop performance.
NOS440:
You are a dumbass. You post to rant and rave at everyone about how fast your P4 is. That's like saying: HA! I got a Ferarri F550, it puts your Pontiac GrandAm to shame. Why don't you just tell everyone how much faster your GF2 Ultra is that a GF2 MX? You have too much time on your hands and not enough other hobbies to spend your paycheck on. I'll never spend $500 on a CPU, well maybe if I made a 7-figure salary. Shut you hole! No one cares!!!!
You are a dumbass. You post to rant and rave at everyone about how fast your P4 is. That's like saying: HA! I got a Ferarri F550, it puts your Pontiac GrandAm to shame. Why don't you just tell everyone how much faster your GF2 Ultra is that a GF2 MX? You have too much time on your hands and not enough other hobbies to spend your paycheck on. I'll never spend $500 on a CPU, well maybe if I made a 7-figure salary. Shut you hole! No one cares!!!!
I would love to see P4 vs. T-bird in REAL multi-tasking OS (eg. some kind of Unix, FreeBSD, Linux or BeOS). Nearest school's server I know is duron 600 with 192Mb of RAM running RH 6.2. It's load is somewhere around 2-3 when powering ~10 X-terminals, sharing disks over several NIC's, running Apache web server and mailserver etc. I'd like to test P4 under same condition to see how multi-tasking powerful it really is. Somewhy I think P4 wouldn't shine on that kind of situation at all ...
<< Hanky I agree But there are no programs that will show us yet what the SSE-2 will have in store for the P-4. Well except the one that Toms Hardware did and if this is any indication the P-4 is going to rock the T-bird world there too !!! But there is no fact to this so lets not slam me for it guys. >>
Well, but I can safely say I will always face problems were I need a strong FPU and SSE-2 will NOT be of any help. Why? Because I'm the one who has to write the programs. I study physics and because computational physics is a must-subject here, I have to do numerical calculations frequently without any program available which could solve the problem. And most of the compilers don't support SSE, 3Dnow! or stuff like that. I now use one brand new version of a compiler and they just implemented MMX (!!!!).
I think a T-Bird helped me more than a P-III or P4...
(Btw: even with my T-Bird it sometimes takes a day or so to calculate specific problems...it took several days with my old K6-2
)<< Well, but I can safely say I will always face problems were I need a strong FPU and SSE-2 will NOT be of any help. Why? Because I'm the one who has to write the programs. I study physics and because computational physics is a must-subject here, I have to do numerical calculations frequently without any program available which could solve the problem. And most of the compilers don't support SSE, 3Dnow! or stuff like that. I now use one brand new version of a compiler and they just implemented MMX (!!!!). >>
Exactly! Since I'm in the graphic business, programs such as Maya, SoftImage, 3DS Max also need Strong FPU. SSE-2 isn't implemented in these programs as well since it requires mostly calculations to render scenes. MMX is implemented and there isn't much gain(since both platforms have it). This is where the 1.2 Tbird shines, 784mb RAM, mounted with a Nvidia Geforce2 Quadro (which is available at work) it performs pretty close if not better than a Duel Xeon 800mhz system(also have it at work) with 1G RAM with the same card. To make it worst, my company can purchase 2 Tbirds 1.2Ghz at the price of 1 Duel Xeon rendering system(this cost ,more actually). I'm quite sure the P4 can't outperform a Duel Xeon system, infact, it does very poorly with calculations compared to the Coppermine.
this is from overclockers.com
CPU and Memory Bandwidth scores at 9x150 = 1.43 GHz.
At 1.43 GHz and a system bus of 150 MHz, the Athlon easily bests the 1.5 GHz Pentium 4 (except for memory). I also ran 3D Mark 2000 and compared the results of 100 versus 133 MHz. There was a definite improvement here, with 100 MHz yielding a score of 7412 while 133 MHz produced an average of 8103 3D marks.
The increased bandwidth definitely allows the system to spread its wings. In addition to 3D Mark, I averaged some runs with the old team Lamb Chop Seti work unit, under client version 3.0. At 100 MHz, completion times consistently came in around 3:38 compared to 3:27 at 133 MHz. The Seti client is very sensitive to memory performance and overall bandwidth; here again, the increased bus speed shows its presence.
CPU and Memory Bandwidth scores at 9x150 = 1.43 GHz.
At 1.43 GHz and a system bus of 150 MHz, the Athlon easily bests the 1.5 GHz Pentium 4 (except for memory). I also ran 3D Mark 2000 and compared the results of 100 versus 133 MHz. There was a definite improvement here, with 100 MHz yielding a score of 7412 while 133 MHz produced an average of 8103 3D marks.
The increased bandwidth definitely allows the system to spread its wings. In addition to 3D Mark, I averaged some runs with the old team Lamb Chop Seti work unit, under client version 3.0. At 100 MHz, completion times consistently came in around 3:38 compared to 3:27 at 133 MHz. The Seti client is very sensitive to memory performance and overall bandwidth; here again, the increased bus speed shows its presence.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter