I tried pushing it harder, but 2.8 is about the best it can do without a vcore increase. 2.9 was no problem at about 1.42V.
I went for the gusto, and 3GHz seems possible, but it's going to need >1.6V. My temps are exceeding 60C while priming in that range, and I just can't maintain stability in those conditions. These Opties run hot.. to say the least..
Messing with my TCCD again, as this memory controller behaves very differently.. many timings need to be adjusted.. I just love new toys
I've had a hell of a time getting my TCCD to behave with this IMC. I'm not sure what it is, but anything above 250MHz just doesn't happen. Going to need to investigate further.. My Winchester behaved in the same manner.
2.9GHz is the real sweet spot with this chip.. 1.46volts is plenty. I just played about 2 hours of Day Of Defeat Source with the below configuration. It actually felt noticably faster than usual.
Mate to get my tccd ram to run over 265-300fsb with my 146 I had to really slacken the timings. What board are you running it in. Also my temps dont go over 50c @1.6v but then mine is on water with 2 rads in the loops. Is yours on air Lemon?
Mate to get my tccd ram to run over 265-300fsb with my 146 I had to really slacken the timings. What board are you running it in. Also my temps dont go over 50c @1.6v but then mine is on water with 2 rads in the loops. Is yours on air Lemon?
Jim, I'm pretty puzzled about the memory behavior.. Even with a 7x multi, and 2.5-4-4-8 timings @ 255MHz, I get constant BSOD failures. I've tried many different Async Latency and Read Preamble combinations, which has not helped either.. It's a DFI Lanparty NF4 Ultra-D, and I'm cooling with an XP90.. Makes me wish that I kept my watercooling These chips run hot, My winchester with 1.48V ran around 44-45 full load, this thing runs about 10C hotter.. I'm thinking about popping the IHS, as some people have had tremendous temperature decreases. I simply can't do anything over 1.55V because of heat..It climbs above 60 very quickly..
I thought I remember hearing that the E6 revision cores were actually dual core chips with one core neutered? Any truth to that Mack?
Can't be..that would mean that the working core isn't centrated under the spreader which would mean warmer temperatures. E6 are cooler also. A lot of people have nuked the heatspreaders on the newer Venice E6 and those only have one core. Venice could be a crippled Opteron with half the cache though.
I have no played with Opteron chips before..... for gaming so I am out of the loop in that aspect. but Im wondering If a Opteron 170 will perform as well as a 4600+ X2 or even a 4800 X2 ? with the cache levels and the voltage I am almost thinking they would perform in the same field!
I have the same stepping as you 543 e4 and my temps are exactly the same I am sure this week runs hotter than the usual I have reset my Big Typhoon 3 times now and I get the same temps, my X2 even runs with less heat at 1.6 than my opty at 1.5.
Sledge, Opteron 170 is a plain dualcore with 2X1024 cache and 10X200. It's somewhere between the 3800 and 4400 in speeds at stock, but they usually have better silicon than the X2's. Still the same core. It isn't faster cause it's called Opteron.
I have the same stepping as you 543 e4 and my temps are exactly the same I am sure this week runs hotter than the usual I have reset my Big Typhoon 3 times now and I get the same temps, my X2 even runs with less heat at 1.6 than my opty at 1.5.
Yep, no doubt about it.. hot week. I have a feeling that removing the IHS would greatly improve the temps. Could be that there is bad contact between the die and the IHS. I'd like to see how it does on a vapo
Yeah pretty much the same by sounds of it I think she would hit 3400 on a vapo quite easily, I think its just the crappy temps keeping us back removeing the hsf could be a idea im considering phase for mine.
I think if it is the HSF causing problems with bad contact this cpu will run at very low vcore and be one of the best yet.
Comments
Single Core. I thought about picking up an Opteron 175, but they were out of my budget I'm afraid.
Just got it installed, so far so good.. no OC yet, but it's coming
CPU-Z 1.31 seems to recognize it properly..
I'm all smiles so far. I havent touched the vcore yet. Memory is running on 2/3 divider for now, will push it higher soon.
Going to let it prime for a bit, and kick it up higher. I havent even found it's maximum at default vcore yet..
Its a rather new stepping.. CABYE 0543FPMW
I tried pushing it harder, but 2.8 is about the best it can do without a vcore increase. 2.9 was no problem at about 1.42V.
I went for the gusto, and 3GHz seems possible, but it's going to need >1.6V. My temps are exceeding 60C while priming in that range, and I just can't maintain stability in those conditions. These Opties run hot.. to say the least..
Messing with my TCCD again, as this memory controller behaves very differently.. many timings need to be adjusted.. I just love new toys
2.9GHz is the real sweet spot with this chip.. 1.46volts is plenty. I just played about 2 hours of Day Of Defeat Source with the below configuration. It actually felt noticably faster than usual.
Jim, I'm pretty puzzled about the memory behavior.. Even with a 7x multi, and 2.5-4-4-8 timings @ 255MHz, I get constant BSOD failures. I've tried many different Async Latency and Read Preamble combinations, which has not helped either.. It's a DFI Lanparty NF4 Ultra-D, and I'm cooling with an XP90.. Makes me wish that I kept my watercooling
I thought I remember hearing that the E6 revision cores were actually dual core chips with one core neutered? Any truth to that Mack?
Memory still on a divider (5/6) until I can figure out why my TCCD is overclocking so poorly..
Look at those load temps
Can't be..that would mean that the working core isn't centrated under the spreader which would mean warmer temperatures. E6 are cooler also. A lot of people have nuked the heatspreaders on the newer Venice E6 and those only have one core. Venice could be a crippled Opteron with half the cache though.
Exactly like an A64 based on the San Diego core (i.e. FX57 etc)
Yep, no doubt about it.. hot week. I have a feeling that removing the IHS would greatly improve the temps. Could be that there is bad contact between the die and the IHS. I'd like to see how it does on a vapo
Did you get similar overclocking behavior to me?
I think if it is the HSF causing problems with bad contact this cpu will run at very low vcore and be one of the best yet.