Got a New Barton 2500+ AQYFA 0342SPMW...Review coming...

[Duvie]

I just got a new Barton 2500+ I am building for a buddie...I am been given free ran to run this up to the limits of the stock heatsink...

The specs are AQYFA 0342SPMW...The heatsink has a bolted on copper core...

My hope is it is multiplier unlocked just incase I can't do 200fsb from stock multiplier. I have pc3200 memory waiting so I am crossing my finger...If it is multiplier unlocekd I don't know if it may do 2.2ghz let alone with this stock HSF hough te cooper core is encouraging....

I am running this on an DFI NF2 ultra.....I will let you know later. His vid card has not arrived so I will have to pull my radeon for time being for testing....

 

[Sarobi]

Most that I've seen have been able to OC reasonably well just by tweaking fsb alone --- w/ locked multi. Be interested to see how this works for you so please let us know. Thanx.

 

[Duvie]

Well it is definitely locked as changing any multiplier in the bios rearding if I wasn't ramping cpu speed would result in me having to clear cmos to get it to reboot...

Right NOw I have it at 1.7v core and 11x200 for 2200 (Barton 3200+)...no problem...starting to stress test as we speak....

Decided to go for the gusto and don't waste my time...Without multiplier cotrols I am either going to swim at 200fsb or sink....Temps look good with idle at 38c with stock heatsink....

 

[Duvie]

So far so good...I have been hitting it wth superpi, POV-Ray, and cinebench and not a hiccup...Will hit the prime95 for the night and then top it off with some tmpgenc testing tomorrow....

Right now it is set at 1.7v vcore bios (don't know if I need less) but this board drop vcore under loading more so then I am used to with my p4....1.7v is reported as 1.69v with CPU_Z with 1.68v reported by hardware monitor. Loading shows drop to 1.6-1.62v...+5v rail and +3.3v rail is strong with 3.3 and 5.03 respectively...

Temps are 34.5c idle as per hardware monitor and 49c load so far in POV-Ray testing....Will see if 6-8 hours of prime will get that into the low 50's....NIce case with excellent case cooling so case temp per digital reading is 76f with room at around 73f...

 

[Jeff7181]

Most likely it doesn't need 1.7 volts... set it back to 1.65 and see if it's stable.

 

[estew33]

Mind if I ask where it was purchased

 

[Duvie]

I got the chip at Newegg.com....


It passed 12 hours of prime v23.7 blended and 10 passes of memtest.....

Starting the tmpgenc test but I think it will be formality....

Here are some numbers

SuperPI

1mb = 0:48 (p4 3.2 = 0:43; p4 3.5 = 0:38)
2mb = 1:50 (p4 3.2 = 1:42; p4 3.5 = 1:34)

POV-Ray v3.5 = 31:43 (p4 3.5 = 26:17 w/ HT 22:40)

P95 v23.7 = 2048K = 154.903 (p4 3.5 = 57.548)

Cinbench2003 single CPU = 269 (329 w/ p4 2.6c@3.2ghz)

TMPGenc (same test as I run with my P4's) = 3:16 (2:08 with 3.5ghz p4)

Besweet ac3enc (same test as I run with my P4) = 4:19 (3:43 with 3.5ghz p4)

Will run some more later.....


System feels really zippy as most amd users attest to, but no more then my 3.5ghz system and the proof is in the benches which I use many real word apps...


I need to tweak memory a bit....Try some cas 2 agressive settings or at least cas 2,3,3,7 like above P4 systems....

 

[Jeff7181]

Pretty good... gettin close to mine, Duvie

SuperPi
1M = 0:45
2M = 1:42

 

[Sarobi]

Very encouraging since I've been unable to find any of these w/ preferred steppings and prior to week 43! Guess I'll have to just take my chances tweaking fsb only like so many others. What kind of Ram are you using?

 

[Jeff7181]

I might be willing to sell my unlocked week 24 to someone if I can get a locked one that's sure to run on a 200 Mhz FSB.

 

[Duvie]

I can't do better then cas 2.5 with these sticks.....Also multipliers are locked cause no matter the setting..i tried 11.5x200 for 2300 t showed up as 2200 in windows....I then bumped the 209fsb for 2300 and no problem at 1.7v...ran a whole bunch of benches but all my ram options go out the door with this board....If I even go to advanced chipset menu in the bios it wills set my fsb back to 133....have auto or spd but can't manual set or go with aggressive cpu and ram timings....

My ram bandwidth increase obviously but my superpi scores do not increase...My tmpgenc time went to 3:10....

I am retesting rpime at 1.675v now and trying to get the best voltage timings and heat package for 200fsb...that is all I was ever looking for....basically give him a 3200+ for the 2500+ money...

 

[Jeff7181]

I bet it's perfectly stable at 1.65 volts.

 

[MDE]

Try setting the memory timings to something like 11-3-3-2 and see if that helps. For some crazy reason the nForce2 likes that high first number (dunno what it's called )

 

[NesuD]

Originally posted by: MonkeyDriveExpress
Try setting the memory timings to something like 11-3-3-2 and see if that helps. For some crazy reason the nForce2 likes that high first number (dunno what it's called )

That Setting is called (Tras) here is what Mushkin has to say about it. This pretty much explains what is going on with that setting and why the higher setting seems to improve mem performance on Nforce2 systems.

Mushkin tRAS explanation

"Memory, in many ways is like a book, you can only read after opening a book to a certain page and paragraph within that particular page. The RAS Pulse Width is the time until a page can be closed again. Therefore, just by definition, the minimum tRAS must be the RAS-to-CAS delay plus the read latency (CAS delay). That is fine for FPM and EDO memory with their single word data transfers. With SDRAM, memory controllers started to output a chain of four consecutive quadwords on every access. With DDR, that number has increased to eight quadwords that effectively are two consecutive bursts of four.

Now imagine someone closes the book you are reading from in the middle of a sentence. Right in your face! And does it over and again. This is what happens if tRAS is set too short. So here is the really simple calculation: The second burst of four has at least to be initiated and prefetched into the output buffers (like you get a glimpse at the headline in a book) before you can close the page without losing all information. That means that the minimum tRAS would be tRCD+CAS latency + 2 cycles (to output the first burst of four and make way for the second burst in the output buffers).

Any tRAS setting lower tRCD + CAS + 2 cycles will allow the memory controller to close the page ?in your face!? over and again and that will cause a performance hit because of a truncated transfer that needs to be repeated. Along with those hassles comes the self-explanatory risk for data corruption. That one is not a real problem as long as the system is kept running but in case it is shut down and the memory content is written back to the hard disk drive, the consequences can be catastrophic. For the drive, that is."

So in theory running 2-2-2-5 is slower than 2-2-2-6. The ideal timings according to Mushkin works out to be:

CAS+tRCD+2 = tRAS

Examples:

2 CAS + 2 tRCD + 2 = 6 (2-2-2-6)

2 CAS + 3 tRCD + 2 = 7 (2-3-3-7)

2.5 CAS + 3 tRCD + 2 = 7.5 (2.5-3-3-8) *Round up*

Apparently with Nforce2 the 200mhz bus sees an added benefit with even higher tRAS than the formula indicates. here is another link to a thread that covers the effect.
AMDMB tRAS thread

 

[MDE]

Woo hoo! I got some awesome backup on that one! Thanks NesuD ! :beer:

 

[Jeff7181]

Originally posted by: MonkeyDriveExpress
Woo hoo! I got some awesome backup on that one! Thanks NesuD ! :beer:

Yeah, that horse is dying =) I don't recall seeing any difference in games a few months ago when I experimented, only in synthetic benchmarks... but hey, every little bit helps right? =)

 

[Duvie]

OK I was finally able to boot with cas 2...I used cas 2,4,4,10 but I may try cas 2,3,3,11 as well...I think that is te best this cas 2.5 rated memory at pc3200 will do anyways...I suggested cas2 stuff but newegg's xmas special and 20+ dollars savings was hard for me to convince him he will feel the cas2....So far in my testing I am not seeing much improvement when I change the cas settings much anyways...

INteresting read I will continue to tweak.....

I may also try 209fsb with cas 2.5,4,4,11 and see if it will hold these settings...

 

[Duvie]

Cannot get any combinations of cas 2,3,3,x..So I will live with the cas 2,4,4,11 which did do better then may cas 2.5,4,4,8 and cas 2.5,4,4,11...I could not get any cas 2.5,3,3,x combination to work as well...

Oh well I am happy with it...I see little to more performance increase going from cas 2.5 to cas 2 in a majority of the test.....I thought the amd setup liked the cas timings more so then my p4 setup??? I get larger leaps going from cas 2.5 to cas 2 on my current p4 setup...


It primed for 2 hours at 1.675v (11x200) so I decided go ahead since the last 2 hours I have been testing povray, superpi, tmpgenc etc, that I should just bump it down to 1.65v and see if she makes it through tonight with the prime95 test....

His vid card has not arrived so some final tweaking is yet resolved.....