Is This Normal?

UnusualSuspect

UnusualSuspect

UnusualSuspect

Associate
Joined
21 Jan 2007
Posts
273
I'm just starting to overclock my C2D by raising the FSB a bit at a time and doing a stress test, as is usual. However, every time I reset it seems to take an awful long time to go from 'black screen' to bootup. Is this normal? Every time I reset I'm waiting up to 10 seconds for the screen to appear - memory check and all that.

I'm using a 2.4GHz C2D on a Gigabyte 965P-DS3 rev. 1. I haven't touched anything else at the moment, so no need to mention other components. Memory is running slightly underclocked atm because of the FSB changes. Currently got the speed up to 2.789GHz.
 
overclock the cpu first.. when u have got the highest stable overclock on the cpu then start upping the fsb.. that way when u hit a wall you will know that it is either the cpu or the ram stopping you...
 
I've been doing that, I was raising the CPU frequency and altering the RAM frequency to keep it equal or under normal. So the CPU speed was the only thing to rise.

But that wasn't my question. It seemed the more I pushed the frequency higher, the longer it took to boot the computer I'd be staring at a black screen for up to 10 seconds before the computer came alive. Is there something wrong, or is this normal?
 
that happens to me too, but i havn't actually waited that long to see if it boots, mine just doesent boot i assume, as my monitor just turns itself off.

Russ.
 
It seems like the startup speed changes only after the BIOS change, because starting it from 'off' it boots up at normal speed. I decided to settle at a frequency of 320. It means my RAM is running at its standard 800MHz and the CPU has gained ~500MHz increase. I took it up to just over 3.2GHz, but I don't like the speed my RAM is running at because of that, which is why I settled at 320.

My 2.4GHz is now clocked at 2.88GHz, which is a nice little boost. Windows Performance Index went from 5.3 to 5.6. :D
 
Just a quick note as we are in the area of strange boots whilst overclocking, i have a 6300+DS3+Geil 2GB PC6400 and have been overclocking and tweaking for the last couple of weeks.
Just upped the FSB to 475 (RAM running at 950) and booted fine, however when i pressed TAB after the restart to check on what speed the CPU is running at and it reported it as 169*7 (1183MHz).
When i get into Windows, CPU-Z reads the correct speed (3325MHZ), has anyone else had this problem?

(soz for the hijack)
 
Unusual Suspect have you tried setting your RAM to a 1:1 multiplier then seeing how fast you can get the CPU? You have to potential to get it much quicker then 2.88Ghz. At 1:1 your ram won't reach its stock speed until a 400FSB. That would give you a nice 3.6Ghz.
 
Darg said:
Unusual Suspect have you tried setting your RAM to a 1:1 multiplier then seeing how fast you can get the CPU? You have to potential to get it much quicker then 2.88Ghz. At 1:1 your ram won't reach its stock speed until a 400FSB. That would give you a nice 3.6Ghz.
Click to expand...
I did have it at 1:1 for a while, just to try and get my FSB up toward 400, but I got an error on Orthos when I reached about 360. I'm not sure if I can go any higher, I hadn't set the FSB and MCH voltages to +0.1v at that point. Does that make a difference with stability? Will it allow my FSB to go higher?

Also, can someone tell me what heats up by changing the FSB speed? Is it the chipset or does it just not cause heat at all? I know its not the CPU because heat changes on that have been minimal at best.

EDIT: I just noticed in the overclocking tips about changing CPU voltage to be able to push the FSB higher. I changed the voltage from stock 2.225 (think that was it) up to 1.3v and the startup time is back to normal. I guess the CPU was having trouble coming to life with such a fast frequency under low voltage.
 
Last edited:
Quick update, cos I'm bored...

I pushed the FSB up until it caused an error in Orthos then raised the CPU voltage a little and carried on. I could get it running stable at 3.4GHz (memory at 760MHz) at a max temperature of 60c. I wasn't entirely happy with 60c, especially as I could feel the heat flowing out the top of my Antec 900. Sort of worrying when you suddenly have another radiator in your room. :p

So for now I'm sticking with 2.9GHz, its fine for me. Maybe in the future I'll get better cooling (using a Freezer 7 Pro atm) and some faster memory so I can switch to 2.5x multiplier for 950MHz, which would be much better.. but 2.9GHz is good enough for now, considering I bought a 2.4GHz. :)
 
You shouldn't need more then 1.45V. Remember to increase the MCH by 0.1V too. That will help a bunch.

Most E6600s can reach 3.6 which would give you stock RAM speeds.
 
Darg said:
You shouldn't need more then 1.45V. Remember to increase the MCH by 0.1V too. That will help a bunch.

Most E6600s can reach 3.6 which would give you stock RAM speeds.
Click to expand...
At 1.35v I was hitting a temperature of 60c, that was with an FSB of 380 so 3.42GHz. I probably would have been able to reach a 400 FSB but I didn't like the hot temperatures, though those temperatures were under a CPU stress test so real world temperatures would no doubt be cooler.

Maybe I'll switch to water cooling sometime in the future and finally find a use for the worst case I've ever bought - the Thermaltake Armour case.
 
You must log in or register to reply here.
Back
Top Bottom