Tualatin 256k better than T-bird?

[Crashman]

Tom just did a comparison of the old Duron to the new Duron (Morgan). What you didn't notice is that the non-overclocked Tualatin 1.13GHz 256k (the wimpy version) did better in most test than the 1.1GHz Athlon and better in half the test than a 1.2GHz Athlon. We already knew the 512k server version offered better performance, but now the 256k version is showing that even the "little brother" Tualatin is beeting the T-Bird.

The thing that makes this test more accurate than the previous overclocked test is that the overcloked test gave the Tualatin the disadvantage of having the bus speed so high as to necessitate using the memory at Cas3 and the AGP at 2x, while the AMD ran at Cas2 and AGP4x.

At stock bus speeds, he was able to run full performance levels for both processors in this latest comparison. Go ahead and start flaming, facts speak for themselves!

I'm so tired of cookies I'd settle for spam!

 

[Crashman]

BTW, Tualatin 256k edging out the T-Bird should position the Tualatin 512k as STOMPING ON the T-Bird and edging out the A4. Perhaps the argument that the T-bird is not really a Pentium III class processor is correct-more like a Pentium 2 class! LMAO@trolls!

I'm so tired of cookies I'd settle for spam!

 

[SerArthurDayne]

um... ok.

/me offers Crashman some nice warm tea to calm him down

"Laziness is a talent to be cultivated like any other" - Walter Slovotsky

 

[Crashman]

Thanks, but I'm a soda drinker. Maybe even Uoisgian Zoda.
I don't even like Intel as a company, I just like the tables to be reversed every few months. And they make the best chipsets in the industry.

I'm so tired of cookies I'd settle for spam!

 

[lhgpoobaa]

so the bus WAS running at 150mhz?

that would have an effect

may have been better to do all the benchmarks with sdr... thus the tully wouldnt have been hundered by its lack of DDR support.

i believe that while the tully is far better performing than the p4, it still has a few negatives, namely:
lack of double pumped fsb/ddr support
based on an ageing design, even though it is 0.13 micron.

other than that its a nice chip. as is the 1ghz morgan...

what i wanna see now is some overclocking on that morgan.
with the pic that tom supplied it looks as if its got its L1 bridges intact (though i could be wrong).
go on tom... overclock the morgan!

P.S. i assume there are no plans for a 133mhz fsb morgan any time soon?

I'll respect your comments & opinions, even if i disagree with them, Provided you dispay maturity.

 

[phsstpok]

I think you're digging here but I'll dig too.

Let's see. So the Tualaton-256, FSB 133mhz, clearly beats a few Athlons, FSB 100mhz, in couple tests but is in line in other tests. It also loses to all Athlons in one test (Dronez). It even loses to the 1000mhz, new Duron, clocked 11.5% lower (1000mhz vs 1130mhz) and despite the FSB disadvantage.

Sorry, I'm not convinced.

 

[Kelledin]

Ahem...cough...eh, the Athlons in this one aren't at their best. The only one at 133MHz FSB is the 1.2GHz, and that one beats the Tually in every test.

Just thought I'd point that out...

Kelledin

"/join #hackerz. See the Web. DoS interesting people."

 

[lhgpoobaa]

mmmm but isnt the athlon/duron/morgan/a4 bus doublepumped regardless of sdr/ddr ram?
what effect does the double pumping have?

i have to admit though i doubt the effect of double pumping the bus can truly be calculated as no single pumped athlons exist... or do they?


I'll respect your comments & opinions, even if i disagree with them, Provided you dispay maturity.

 

[Crashman]

FSB disadvantage? What about AMD's glorious "200MHz FSB"? And even Tom pointed out that the Dronez test was a anomoly, perhaps it's AMD optomized only? Anyway, a standard method for doing statistics is to throw out the one highest and one lowest score to avoid such anomolies when comming up with a general conclusion. Anywho.....I'll try to remember it if I ever decide to become a professional Dronez player!

I'm so tired of cookies I'd settle for spam!

 

[phsstpok]

Yup, that's the hype.

I also have doubts of the effectiveness when the northbridge is limited to the FSB's base clock speed at least with 100 mhz Athlons with early chipsets.

 

[Crashman]

Here's my situation:
1) If I can't get a Tualatin up to 2GHz I won't be upgrading to it. I already have a PIII 700@933 and am looking for a MAJOR upgrade in about 6 months. Already considering it.
2) If the Northwood doesn't kick the crap out of the Tualy I won't even consider it. If it does, at a reasonable price point, I'll consider it.
3) If I find a motherboard that meets my satisfaction for the newest Athlon, I will be switching over. ALi ALMOST qualifies with their chipset, but with no motherboards that meet my level of inspection (yet). And the nForce and 735 are not yet proven to my satisfaction. So only time will tell.
4) I might just get a voltage adapter and throw a Tualatin on one of my current boards if solutions 1-3 are not viable.

I'm so tired of cookies I'd settle for spam!

 

[Crashman]

Um,,, er,, Say, wasn't that 134 for the 1.13 Tualy compared to 1.30 for the 1.2GHZ T-Bird in internet content creation? And 168.1 for the Tualy, 163.5 for the 1.2G Bird in Q3A? 2 out of 5 ain't to shabby, especially considering it's lower clock rate and 1/2 the bus speed, according to AMD 133 vs. 266! LOL!
BTW, when you consider that the Athlon has DDR FSB, isn't the memory speed the true place to compare (133 vs. 133)? OUCH, forgot that the Athlon had DDR RAM also! Geeze, and the Tualy still beet it clock-for clock! That's gotta hurt, especially when you consider that a different version of the Tualy is also avialable with TWICE THE CACHE OF THE ONE TESTED!

I'm so tired of cookies I'd settle for spam!

 

[phsstpok]

(See my last post).

Anyway, a standard method for doing statistics is to throw out the one highest and one lowest score to avoid such anomolies when comming up with a general conclusion

OK, I'm still playing. Let's throw out the Dronez benchmark but then we must also throw out Tualatin's best result which would be the Q3A Demo1 test.

That leaves

<b>Sysmark2001</b>

Tualitin 1.13 - 128, Athlon 1100/100 - 121, Athlon 1200/100
In line, no winner

<b>Symark2001 Internet Content Creation</b>

Tualitin 1.13 - 134, Athlon 1200/100 - 128, Athlon 1200/133 - 142
not inline for 100mhz Athlons but not conclusive for 133mhz Athlons. I'll give you this one.
Score 1 for Tualitin.

<b>Symark2001 Office Productivity</b>

Tualatin 1.13 - 123, Athlon 1100 - 128, Athlon 1000 - 120
Score 1 for Athlon

<b>Quake3 Arena NV15</b>

Tualatin 1.13 - 37.7, Athlon 1100/100 - 38.1, Athlon 1000/100 - 35.4
Score another 1 for Athlon.

<b>Total</b>
Athlon 2, Tualatin 1, Draw 1, Tests thrown out 2

<i>Tualatin 256k better than T-bird?</i> - Crashman

I'm still not convinced.

 

[dhlucke]

Kinda inconclusive. They're all close. I wouldn't look twice right now. Take a closer look in 4 months when that 6 month mark is coming up.

<font color=red>Change the sig of the week!</font color=red>

 

[Crashman]

OK, but even if the 256k is almost tied with equal speed T-Birds, where does that place the 512k? And maybe Tom should have run some more test, such as 3DM2k and 2001, etc, just to give us a clearer picture?
Anyway, I started this post to bring out the trolls, were are they? I love arguing with the trolls, it's so easy!
As previously stated, if I see a significant move in the motherboard industry, I might move to Athlon anyway, it's cheap. If I see a significant improvement in the P4, I could go that way, but it's not too promising. Otherwise, I'm more likely to simply go with a Tualatin and a voltage converter on one of my current platforms, simply so that I can keep my wonderful Intel chipset. With the converters comming out, I doubt I'd switch to a new platform for the Tualatin. Those AMD prices......well, that's what got me into my VIA mess to begin with, which lead to my Intel conversion. But being the cheapskate I am... CAN'T ANYONE MAKE AN AMD BOARD AND CHIPSET AS GOOD AS THE CUSLS/i815E?

I'm so tired of cookies I'd settle for spam!

 

[phsstpok]

Yes, I mostly agree, more tests. An Athlon 1000/133 would have rounded out the tests a little better. However, what would be much better would be a comprehensive, head-to-head, multiplatform comparison. It should take into account CPU and Graphics scaling (for the masses) but but also have good sidenotes on specialty applications. It's just time for all the information to be put in one place. (Tom are you listening?)

Personally, I prefer AMD processors but that's because they have always been cheaper than Intel (sometimes much cheaper) but I am getting a little tired of VIA quirks [understatement].

 

[Crashman]

I knew it was too close to call, at least conclusively (one side could show one test, then another the other, until they were blue in the face, if there were enough sites with enough benchmarks). But I was hoping to have broiled troll for dinner!

I'm so tired of cookies I'd settle for spam!

 

[Crashman]

You have to remember that I also use my system to test parts! So my former VIA problems were of great significance! AMD761 takes care of the AGP problems associated with VIA, but the use of the VIA 686B southbridge leaves much to be desired. The ALi chipset looks good, especially revision B0, but even the best board that uses it (A7A266) has minor problems. nForce and 735 are unproven....too much going on for anyone with a decent system to even consider doing an upgrade right now, except for maybe a processor swap.

I'm so tired of cookies I'd settle for spam!

 

[Kelledin]

hmmm maybe the trolls have gotten smarter. or maybe the AMD trolls are getting more proactive than reactive. Oh well...

Kelledin

"/join #hackerz. See the Web. DoS interesting people."

 

[74merc]

<i>oh yea, well... well, uh... AMD rocks and Intel sucks and you're a doodiehead!!</i>

it was really too close to call, I noticed how well the tually did, I'm impressed.
where, oh where was the P4 in those benchmarks, huh?
I'm with you, I want a tually @ 2Ghz... lessee what the old Pentium Pro can do against some 6(is that right?)year younger processors...
one way or another, I will have a Tually @ 1.4Ghz in my laptop, even if it requires a motherboard upgrade... may go higher than that if possible, but it seems that will probably be the end of the mobile PIII's...

----------------------
Independant thought is good.
It won't hurt for long.

 

[Matisaro]

Nice, cant wait till everyone is on the .13 micron process. I will take the 1.4ghz tbird for half the price though .

PS: I am not anti intell, I am anti p4(in its current form).

~Matisaro~
"Friends don't let friends buy Pentiums"
~Tbird1.3@1.55~

 

[AMD_Man]

Ok, the T-bird was at a slight disadvantage...it was using a 100MHz bus with PC133 RAM. The double pumped 100MHz bus is not taking full advantage of the low latency of PC133. Therefore, if you wanted to compare the Tully and the Athlon, you would use a 133MHz bus and PC133. A board based on the KT133A would be best.

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor

 

[zengeos]

I haven't looked at the comparison as yet, but isn't it a rather moot comparison since TBird is being phased out and since Palomino will start at 1.3ghz (I think that's the low clockspeed on release) Also, just like Tualatin has some significant improvements, not the least of which is .13 micron, Palomino also has some significant improvements (not .13 micron though) which add up to at least a 5% performance improvement. So, while it's an interesting comparison it seems a little moot with the impending release of Palomino on the desktop.

Mark-

When all else fails, throw your computer out the window!!!

 

[camieabz]

Q. Will the 512K Tualatin be affordable before Intel gets the P4 & Celeron bus rolling?

<b><font color=blue>Change the</font color=blue><font color=red> sig of</font color=red><font color=green> the week!</font color=green> :tongue: </b>

 

[camieabz]

Q. Will the 512K Tualatin be affordable before Intel gets the P4 & Celeron bus rolling?

<b><font color=blue>Change the</font color=blue><font color=red> sig of</font color=red><font color=green> the week!</font color=green> :tongue: </b>