• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Any news on HD2950XT?

As far as im aware these 2x cards are essentially just what we have now from ATI, but changed to a 55nm process, and also have their bus's cut to 256bit, so why would you want to get rid of a 2900 XT for another 2900 XT, but on a smaller process, and a bus cut, or even a 2900 pro for another 2900 pro on a smaller process with a bus cut, makes no sense at all, and also the Dx10.1 is pointless.
but hopefully it'll be cheaper for the people who didn't buy a 2900xt/pro :D
 
It wasnt as if they were bandwidth limited though.

Agreed they had more bandwidth than a gtx but was slower, so i would think there is some head room in the bandwidth to cut it down if it means a cheap card at about the same speed to go against nvidia then that what they will do.
 
the 2600 haven't got voltage controls. i think it does it by itself..

anyway since i've overclocked the cpu to 3.4gz. it seems to be a different now at 884 its added 2 fames extra

Ahh shame :(. I think the progress you are seeing is from increased sm2/sm3 scores which is happening with the same overclocks on the GPU but faster CPU speed. It seems we are CPU bound a little in comparison to Conroe systems as if you had a Conroe at lets say 3.4Ghz then your SM2 and SM3 scores would be better at the same GPU overclock. I think if you had access to voltages that a core increase would improve scores.
 
ive got my cpu to 3.45ghz at 47c, but i don't know what max temps this cpu can take. also i don't know if 47c is 1 of the cores or the whole cpu. the bios just says cpu temp 47c... whats a safe temp on both core for my cpu?
 
I think it's 71'C in Coretemp and that usually reports back about 50-55'C in programs like Everest/Speedfan etc. This was on a much older version of coretemp and also on the 939 platform in my case.

When I run a CPU stress test if my CPU goes above 71'C or 72'C the stress testing fails on one of my cores and it's heat that causes my stability. One of my cores is a beauty while the other one is an ugly sister.

What I'd do is just keep messing about with it until 3dmark crashes on the CPU test. That would mean that you are just past your limit so you could just drop back to the last configuration that worked and stick at that until you do an 8 hour stress test. Do not go above 1.55V though. You could do a benchmark at 1.6v but the gains would probably be minimal and the heat would be too much (most likely). After you're finished with the benching side of it then do an 8 hour stress test while you sleep. This way you'll know what's stable for 24/7 use.
 
I think it's 71'C in Coretemp and that usually reports back about 50-55'C in programs like Everest/Speedfan etc. This was on a much older version of coretemp and also on the 939 platform in my case.

When I run a CPU stress test if my CPU goes above 71'C or 72'C the stress testing fails on one of my cores and it's heat that causes my stability. One of my cores is a beauty while the other one is an ugly sister.

What I'd do is just keep messing about with it until 3dmark crashes on the CPU test. That would mean that you are just past your limit so you could just drop back to the last configuration that worked and stick at that until you do an 8 hour stress test. Do not go above 1.55V though. You could do a benchmark at 1.6v but the gains would probably be minimal and the heat would be too much (most likely). After you're finished with the benching side of it then do an 8 hour stress test while you sleep. This way you'll know what's stable for 24/7 use.

My cores ain;t happy if they go over 53 degrees :eek:

And one on mine is a cow. Only one gets to 53 and the other sits happily at 43. That's what's stopping me get past 3.25Ghz :(
 
My cores ain;t happy if they go over 53 degrees :eek:

And one on mine is a cow. Only one gets to 53 and the other sits happily at 43. That's what's stopping me get past 3.25Ghz :(

[EDIT] I thought you had AM2 for some strange reason. That's what happens if your typing, on the phone and having a herbal roll up at the same time :o. Was this using Coretemp though as that is way too low to fail as 939's have to hit past 71'C in the core to fail a stress test IIRC?. Shame to hear about the naff core though. I think AMD done this on purpose so everyone wouldn't buy lower versions :rolleyes: [/EDIT]



Ahhh. So these have lower volts than 939 and a lower temp cut off point because they are cooler?. You are talking about Coretemp though?.

As AM2 wasn't up to much over 939 I haven't really learned much about these CPU's.

What voltage do you run for stock clocks Gareth?.
 
Last edited:
Ahhh. So these have lower volts than 939 and a lower temp cut off point because they are cooler?. You are talking about Coretemp though?.

As AM2 wasn't up to much over 939 I haven't really learned much about these CPU's.

What voltage do you run for stock clocks Gareth?.
1.40v

Field Value
CPU Properties
CPU Type DualCore AMD Athlon 64 X2, 3000 MHz (15 x 200) 6000+
CPU Alias Windsor
CPU Stepping JH-F3
Instruction Set x86, x86-64, MMX, 3DNow!, SSE, SSE2, SSE3
Original Clock 3000 MHz
Min / Max CPU Multiplier 4x / 15x
Engineering Sample No
L1 Code Cache 64 KB per core (Parity)
L1 Data Cache 64 KB per core (ECC)
L2 Cache 1 MB per core (On-Die, ECC, Full-Speed)

Multi CPU
Motherboard ID TEMPLATE
CPU #1 AMD Athlon(tm) 64 X2 Dual Core Processor 6000+, 3006 MHz
CPU #2 AMD Athlon(tm) 64 X2 Dual Core Processor 6000+, 3006 MHz

CPU Physical Info
Package Type 940 Pin uOPGA
Package Size 4.00 cm x 4.00 cm
Transistors 227.4 million
Process Technology 9Mi, 90 nm, CMOS, Cu, DSL SOI
Die Size 230 mm2
Core Voltage 1.100 - 1.450 V
I/O Voltage 1.2 V + 2.5 V


CPU Utilization
CPU #1 / Core #1 6 %
CPU #1 / Core #2 20 %

Field Value
Sensor Properties
Sensor Type ITE IT8716F (ISA E80h)
GPU Sensor Type Diode (ATI-Diode)
Chassis Intrusion Detected Yes

Temperatures
Motherboard 40 °C (104 °F)
CPU #1 / Core #1 37 °C (99 °F)
CPU #1 / Core #2 36 °C (97 °F)
Aux 46 °C (115 °F)
GPU1: GPU Diode 58 °C (136 °F)
GPU2: GPU Diode 55 °C (131 °F)
Hitachi HDS722512VLSA80 34 °C (93 °F)
Hitachi HDS722516VLSA80 32 °C (90 °F)
Maxtor 6V160E0 37 °C (99 °F)
WDC WD2500KS-00MJB0 39 °C (102 °F)
WDC WD3200AAKS-00SBA0 34 °C (93 °F)

Cooling Fans
CPU 2961 RPM
Chassis 1870 RPM
Power Supply 1418 RPM

Voltage Values
CPU Core 1.40 V
+2.5 V 2.99 V
+3.3 V 3.23 V
+5 V Standby 5.13 V
VBAT Battery 3.25 V
 
Last edited:
http://www.hothardware.com/articles/AMD_Athlon_64_X2_6000/?page=2

http://www.atomicmpc.com.au/article.asp?CIID=76767

http://www.firingsquad.com/hardware/amd_athlon_64_x2_6000/

http://www.xbitlabs.com/articles/cpu/display/athlon64-x2-6000_8.html

All of these are reporting 3.2Ghz for their max OC ;). Either you have a great new stepping or your lucks on the up.

I think it's just the same as in overclocking 939 so just keep it under 70'C through Coretemp. Temp reading programs that do not do the core will report back roughly about 50'C - 60'C but coretemp is the only reading you need.

Set it at the speed and voltage you want and run a 8 hour stress test (small FFT's). If a core gets too hot it will fail the core anyway so you won't wake up to a melted PC :D.

What I do is sit and keep my eye on it for the first hour. If the temps are under 70'C after the hour then it's more than likely to make it through the night and I just awaken or return to see what's happened. If it failed I would set it back to a working overclock for that day and then do another stress test when I sleep again.

I'd stick to the volts I mentioned until you feel more confident. 1.55vmax for now(If you ever need it).
 
Back
Top Bottom