p3/667e as fast as a t-bird 900?

Nexus7

Distinguished
Jan 16, 2001
31
0
18,530
I'm running two systems, an t-bird 900 machine with an MSI k7tpro2a board and a p3/667e on an ABIT BE6 board. Both machines have the same hard drive (WD 18gig ata66 7200rpm) and the same video card (Leadtek Geforce 2 32/ddr). I get the exact same performance in UT and in Giants on both systems, yet the P3 system seems to be alot more stable. I don't the the freeze then go FPS on the P3 but it happens alot on the t-bird system. Both boards have the latest bios and hardware updates. Is this common? I thought the Athlon was a better and faster chip? Could it be the VIA chipset that I'm running? I've heard many problems with VIA. Any advice will help.
Thanks
 
G

Guest

Guest
interesting...clock for clock, the p3 is the better chip. ive found p3's generally run as fast as the equivalent t-bird that runs about 50mhz faster. durons more like 100mhz...i work at a computer store so ive had plenty of time to test this sort of stuff. anyway, thats the results ive got. if anyone has found different results then post away, i'd be interested to hear what others have found.

oh yeah, if youre about to flame me for not supporting amd every chance i get, my flagship system is a duron 800@1016, im just telling you the truth here.....
 

toonces

Distinguished
Dec 31, 2007
213
0
18,680
not to sound sceptical or anything, but let's see some benchmark scores. and the complete listing of components in your systems.

<font color=red>booyah, grandma, booyah..</font color=red>
 
G

Guest

Guest
what OS are you running?

M

one of the first UK T-Bird users....
 

Sojourn

Distinguished
Dec 31, 2007
131
0
18,680
For the different results you are seeking you could refer to this very website. Even with possible chipset performance issues, the Athlon Thunderbird is very comparable clock for clock to a pIII, each doing slightly better at certain tasks. The pIII certainly doesn't hold a 50MHz performance edge. In fact, the pIII had only about a 50MHz perfomance lead over the Athlon Classic due to its off die cache running at a fraction of the CPU clock, severly crippling its performance. The Thunderbird core solved this problem and put the Athlon on equal footing with the pIII. Tom's benchmarks (as well as other websites) have shown all of this. Do you frequently run benchmarks at your computer store or are you simply "eyeballing" this supposed performance edge?

As for a pIII 667 matching an Athlon 900MHz, I don't think so. Either Nexus7 doesn't know how to assemble a PC or he's a troll looking to start a flame war. Check out any of the dozens of benchmarks available here or on other web sites to see what sort of performance you should expect with each system.
 

Nexus7

Distinguished
Jan 16, 2001
31
0
18,530
I'm not trying to start a flame war here. I have built several systems in the past with pretty decent hardware. While the Athlon T-bird is a fast chip accourding to online benchmarks that does not explain why I get silky smooth framerate on UT and Giants (98se) on the p3 and choppy fps on the athlon. Both machines are running the 6.31 det.
I really like my Athlon machine. Do you have two machines side by side to compare to see or are you saying this based on other peoples benchmarks. In Sandra benchmarks my Athlon thrashes the P3, but in real world gaming they are the same from what I'm seeing. I'm not an Intel or Athlon basher, I have both. This was just a question to see if anynoe else had similar problems. Maybe it's something I overlooked, a driver, a patch......I'm not sure. I don't come to these forums to bash people, I come to ask questions or maybe help someone.

Nex
 

Nexus7

Distinguished
Jan 16, 2001
31
0
18,530
I'm not too sure how to post an official benchmark, but I'll try. As for my components here goes

For the Athlon

MSI k7t pro2a with Via 4.25a and latest bios
T-bird 900 with super orb cooler
300 watt power supply
128 meg CAS2/133
Leadtek Geforce2 DDR 32meg with new bios
SB live 5.1
18gig ata66 7200rpm HD
48x creative cdrom
win98se

For the Intel

P3/667 with golden orb cooler
Abit BE6 with last bios
300 watt power supply
128 meg CAS2/133
Leadtek Geforce2 DDR 32meg with new bios
SB live 5.1
18gig ata66 7200rpm HD
48x creative cdrom
win98se
 

tfbww

Distinguished
Jan 3, 2001
211
0
18,680
I think even THG has shown that "Clock for Clock" the P3's outperform the Tbirds. The url below, for instance, shows that a 1G P3 gets 154 fps in Q3 versus 145 for the Tbird 1Gig. The results are similar with the 933/950 MHz chips and in UT. The difference, of course, is the $150 in the "clock for clock" comparison. THG in the below review did a price:performance ratio (pre-price cut): P3/1G 18fps per $100 and the 1G/Tbird was 32. I think the price cut was around 30-40% so that means the new number would be 23-25 for the P3.

http://www.tomshardware.com/cpu/00q4/001017/athlon-10.html

(PS, this is not a slam of either Intel nor AMD. Anyone misconstruing it as such should stick to the flamefest in one of the other AMD/Intel bashing sessions.)
 

smn198

Distinguished
Dec 31, 2007
179
0
18,680
Easy. You are running at a high resolution. The graphics card is the limiting factor so CPU speed will have little effect. You may want to install the VIA 4-in-1 drivers to help the athlons stability.

<A HREF="http://www.anandtech.com/mysystemrig.html?id=1686" target="_new">System spec.</A> Ideas appreciated.
 

Sojourn

Distinguished
Dec 31, 2007
131
0
18,680
Clock for Clock, the pIII's beat the Athlon in some benchmarks, the Athlon wins in others. Generally the pIII does better in memory intensive benchmarks like Q3 due to VIA's subpar memory performance. The Athlon dominates any FPU dependant benchmarks. pIII performs better in q3 and UT, the Athlon performs better in Linux kernal compilation and 3D Studio Max (even a 900MHz Athlon beating the OC'd 1072MHz pIII by 20%.)

Since that article the pIII 1GHz has dropped in price by 40%, the Athlon by 33%. The 1GHz pIII is still more expensive than the 1.2GHz Athlon, despite Intel's deep price cuts, so the Athlon retains its price/performance crown easily. Regardless of this, neither processor domininates the other widely enough to be declared a better performing processor, clock for clock.
 

Sojourn

Distinguished
Dec 31, 2007
131
0
18,680
You won't notice the difference between 90fps and 150fps, especially if your monitor is running at a refresh rate of, say, 75Hz. Browse through Tom's motherboard reviews for benchmark setups to find what drivers and BIOS revisions he uses to get his marks. And yes, I have been to LAN parties with a 1GHz Athlon and pIII running side by side, and they were indistinguishable in performance.
 
G

Guest

Guest
try benchmarking at 640x480 so that the video card isn't limiting your framerates.
 

toonces

Distinguished
Dec 31, 2007
213
0
18,680
go to www.madonion.com and go to their gaming section and download a version of 3Dmark2000. this is the general standard benchmark for gaming rigs. if all things are the same (except for mobos and cpus) then you should get a pretty good measurement, since one of the tests is using only cpu power.

<font color=red>booyah, grandma, booyah..</font color=red>
 

necroscope

Distinguished
Dec 31, 2007
219
0
18,680
my experience has been that they are about the same speed, except in games and SETI. These are what matter most to me.

AMD is better in office appz, but the only office appz I use are Office (frontpage, excel (hey i made tom's seti stats!,powerpoint)

Then the price/performance comes into play and AMD whoops down Intel.
 

Oni

Distinguished
Dec 31, 2007
880
0
18,980
well make sure you have both mother boards with the newest bios/drivers and make sure you defragment both hard drives. If one drive has a cleaner install it could run better than the other. if your Tbird 900 is fragmented it could deffinitely cause problems. I normally get 99.99 fps with my Tbird 700 in half life engine games, and one day I was suddenly getting only 16 fps. SOOOO I ran scandisk and defragged and sure enough 99.99 fps was back.
 

Nexus7

Distinguished
Jan 16, 2001
31
0
18,530
This is what I got from 3dmark2000.

On the AMD Machine on 1024x768x16
6812 marks

On the Intel Machine with the same settings
5435 marks

Both machines had the HD defraged prior to the test.
Once again the AMD machine thrashes the Intel one, but the choppyness is still happening on the AMD machine.
I thank you all for your input, I'll continue to use the AMD machine for the time being.
 

jclw

Distinguished
Dec 31, 2007
1,255
0
19,290
My PIII-500@600 (katmai core) takes 9.5hrs (570mins) to finish a SETI workunit (3.0.3 command line). My TBird-900@950 takes 7hrs (420mins) with the same OS (win2000proSP1) - both have 128mb ram and similar configurations.

Dividing that down I would think a 100MHz PIII would take 6 times as long which would be 57hrs (3420mins) and a 100MHz TBird would take 66.5hrs (3990mins). That means the PIII would be 16% faster at completing work units if the chips were running at the same speed. This is despite the fact that the PIII's memory bus is running at 120MHz, and the TBird's memory bus is running at 140MHz.

Running all my other apps show no real overwhelming difference in speed between the two computers. I'm guessing a PIII-866EB whould be a good match against a TBird-1000 (and the same price) for what I do.

-JW
 
G

Guest

Guest
Obviously you have done something wrong. Try to reconfigure the system properly. Get the latest drivers etc. If what you posted is true(which I doubt).

I don't think that all the Hardware Sites on the internet would have missed something of this magnitude. Ya, right a P3-667 beating a t-bird 900. Intel would have a field day with this if it were true.

I personally think you are trolling and if your not I have a really nice piece of land to sell you. It's in a beautiful country with a rich history. You might know it as Atlantis. Beach front property. What a deal, act now.

Note: To the other forum members: Remember the bigger the lie the faster the morons in society will believe it. Look at the published facts this guy is a troll or a very inept box builder or both.
 

Nexus7

Distinguished
Jan 16, 2001
31
0
18,530
Why don't you freaking read the rest of the replies in this post before making such comments. I showed my 3dmarks and it clearly states that the Athlon machine beats the Intel. My question has been that in UT and in Giants the Intel gives me smoother FPS under similar conditions minus the board and processor. The P3 and Athlon for SOME reason had the same FPS in UT, that is why I created this post. Not to troll around or start a flame war or to be told that I can't build a system properly. Since the time that I posted this message I've gotten a few tips to check on my system. If you don't have anything nice to say, keep it to yourself. I come here to ask questions and get answers, not to get into verbal arguments with know it alls like you. Go find another post to flame on.
 
G

Guest

Guest
.... There were problems with some Athlon MoBos and AGP GeForce 3D cards. It made the 3D slow and choppy. I think the updated drivers are now available to fix that.

.... Also, the monitor refresh rate auto synchronization must be turned "OFF" to run benchmarks, otherwise, nothing runs higher fps than the monitor's auto refresh rate!!

.... Generally the Athlon 900 should thrash the PIII 667 by about 40-100%.
 

DSutcliffe

Distinguished
Dec 31, 2007
183
0
18,680
I had a quick question, how do you test FPS in UT when you are playing? or do you have to run the demo?

T-Bird 800Mhz
2X IBM 75GXP 30GB in RAID0 config
Geforce2 32MB
SB Live
 

flavio321

Distinguished
Jan 29, 2001
321
0
18,780
i don't know about ut but in counterstrike, you bring down the console and type cl_showfps 1


If you can't beat 'em kill 'em
athlon "SLOTA" thunderbird 700@1050mhz
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
This brings up an interesting possability to theorise about.

Every benchmark score I've ever seen is with sound turned off or more often with a soundcard not even installed. And without a modem, network card, or any other unneeded card installed.

But what if the PCI systems could actually serve to slow down a VIA motherboard more than they do an Intel motherboard. Then, in theory, for actual gaming with sound and network play involved, an Intel chipset motherboard and Intel CPU might noticably out perform a VIA chipset motherboard with an AMD CPU.

Keep in mind that this is all theory. So don't anyone flame me for coming up with the concept. But we already know that Intel motherboards have an advantage over VIA motherboards, so I'm only expanding upon that.

Does anyone know of any benchmarks that actually compare complete systems with at the very least sound turned on?

- Sanity is purely based on point-of-view.