- Joined
- Apr 10, 2001
Well, got my TBRED B 2100+ from Newegg yesterday. They really are great, got it in 3 days and was as advertised AIUHB.
Plopped it in the KT7A-R. I'm hoping this will tide me over for another 6 months. This chip is replacing a T-Bird 1.4Ghz OC'd to 1.6. I have a custom built water cooling unit to cool the chip and the KT7A-R has been voltage modded with the micro potentiometer allowing voltages to a little over 2.01 Volts.
I set the board back to stock voltage and began testing. I came up with about the same results as from THIS THREAD .
The multipliers I got pretty much matched the following from the KT7 thread:
500 -> no boot
550 -> no boot
600 -> no boot
650 -> no boot
700 -> 1500
750 -> 2200
800 -> 1600
850 -> 1650
900 -> 1700
950 -> 1800
1000 -> 2300
1050 -> 2400
1100 -> no boot
1150 -> 1900
1200 -> no boot
>= 1300 -> 2000 (I think the >=1300 setting corresponds to the 12.5 multiplier.)
1400/133 -> no boot
Although my differences were:
500 = 2000
1133(133) = no boot
1200(133) = no boot
1333(133) = no boot
1400(133) = no boot
One thing I did notice, when the voltage went over 1.85V the system would not boot. I'm assuming that is some safety feature of the 2100+. I've not looked at the spec sheets yet so I don't know what the max voltage is. If anyone can answer that question it'd save me the time of trying to find out.
Prime 95 would not run at any speed on my system. Although the system was running stably, as soon as I lauched Prime 95 it would lock up. Sisoft Sandra 2003 burn in would work fine, however, as well as all the benchmarks.
The max speed I attained was 2400 (1050 in BIOS), but the system wasn't stable. Temps were not the problem as I never saw it go over 40C with a reading from a digital thermometer.
Currently the system is running at 2300 (1000 in BIOS) and I've set the DRAM speed to be HSTCLK+PCI to get the RAM to run at 133 rather than 100.
I'm still not sure it's 100% stable, but I plan to continue testing.
I will say this....the system runs about 3x as fast as it did with the 1.4Ghz T-Bird. It's definitly a very noticable improvement in speed and Sisoft Sandra Benchmarks are about double or a little more what they were with the 1.4 T-Bird.
Using manual settings also worked, although using manual settings you won't get the full speed of the chip as there is no multiplier of 23. I believe the highest is 14 if memory serves so you have to use the standard settings if you want to achieve 2300Mhz.
When I have some free time I'm going to go back and double check the settings that gave no boot. I had set the voltage to 2.0 V while testing those and I believe that may have affected the results as the system won't boot not matter what speed you set it to if the voltage is over 1.85 actual.
Comments?
Plopped it in the KT7A-R. I'm hoping this will tide me over for another 6 months. This chip is replacing a T-Bird 1.4Ghz OC'd to 1.6. I have a custom built water cooling unit to cool the chip and the KT7A-R has been voltage modded with the micro potentiometer allowing voltages to a little over 2.01 Volts.
I set the board back to stock voltage and began testing. I came up with about the same results as from THIS THREAD .
The multipliers I got pretty much matched the following from the KT7 thread:
500 -> no boot
550 -> no boot
600 -> no boot
650 -> no boot
700 -> 1500
750 -> 2200
800 -> 1600
850 -> 1650
900 -> 1700
950 -> 1800
1000 -> 2300
1050 -> 2400
1100 -> no boot
1150 -> 1900
1200 -> no boot
>= 1300 -> 2000 (I think the >=1300 setting corresponds to the 12.5 multiplier.)
1400/133 -> no boot
Although my differences were:
500 = 2000
1133(133) = no boot
1200(133) = no boot
1333(133) = no boot
1400(133) = no boot
One thing I did notice, when the voltage went over 1.85V the system would not boot. I'm assuming that is some safety feature of the 2100+. I've not looked at the spec sheets yet so I don't know what the max voltage is. If anyone can answer that question it'd save me the time of trying to find out.
Prime 95 would not run at any speed on my system. Although the system was running stably, as soon as I lauched Prime 95 it would lock up. Sisoft Sandra 2003 burn in would work fine, however, as well as all the benchmarks.
The max speed I attained was 2400 (1050 in BIOS), but the system wasn't stable. Temps were not the problem as I never saw it go over 40C with a reading from a digital thermometer.
Currently the system is running at 2300 (1000 in BIOS) and I've set the DRAM speed to be HSTCLK+PCI to get the RAM to run at 133 rather than 100.
I'm still not sure it's 100% stable, but I plan to continue testing.
I will say this....the system runs about 3x as fast as it did with the 1.4Ghz T-Bird. It's definitly a very noticable improvement in speed and Sisoft Sandra Benchmarks are about double or a little more what they were with the 1.4 T-Bird.
Using manual settings also worked, although using manual settings you won't get the full speed of the chip as there is no multiplier of 23. I believe the highest is 14 if memory serves so you have to use the standard settings if you want to achieve 2300Mhz.
When I have some free time I'm going to go back and double check the settings that gave no boot. I had set the voltage to 2.0 V while testing those and I believe that may have affected the results as the system won't boot not matter what speed you set it to if the voltage is over 1.85 actual.
Comments?