• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

intel Speedstep causing crashing when on battery after cpu upgrade - E1505

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Kingfish999

Member
Joined
Nov 15, 2010
Location
Palm Harbor
i have an old Dell Inspiron E1505 that a couple years ago i upgraded the CPU from an crappy T1300 single core 1.6ghz to a decent T7600 dual core 2.3ghz. the CPU swap works great and is much faster but always had a crashing issue unless i had it plugged in. if i tried to run on battery it will shortly freeze or crash, then endlessly restart durring POST. i thought it was crappy batteries so i bought 2 new ones and no difference. since then i always used it plugged in because i very rarely need it

here is the thread when i did the swap
http://www.overclockers.com/forums/showthread.php/757234-old-Dell-E1505-cpu-upgrade

well with hurricane Irma knocking out power since sunday (it being wensday now), i have been using it alot. i decided to try and figuer out this issue and narrowed it down to when i disable Intel Speedstep in the BIOS, the issue goes away. i can keep it on battery for 2 hours no problem. but its also slow as heck because its now only at 1.0ghz instead of 2.3ghz. it says right in the description that disabling it will utilize the cpus lowest clock speed. is there a way i can leave speedstep disabled but keep it at the proper clock speed? there msut be some incompatibility with speedstep and this CPU and the way the laptop saves power when unplugged

any ideas what i can do? heres the benchmark with speedstep on and off while plugged in. its at less than 1/2 speed in every test
cpu.JPG
 
theres no changeable settings for the cpu or memory. but it does display correct info for the cpu. i think i did update the bios after the cpu upgrade too. im thinking im kinda screwed unless theres a modded bios avalible
 
Is it downlclocking the memory or the multi ? ( im guessing multi ) maybe a bsel mod ?

its locked the mult. CPU-Z shows bus speed of 166mhz and multiplier of 6 to 14. 6 being 1ghz and 14 being 2.3ghz. putting stress on it does not change the multiplier, its locked at 6.

never heard of a 'bsel mod' till now. sounds interesting, ive done the tape mod to use a Socket 771 Xeon in a Socket 775 motherboard and seems similar idea. anywhere i can get more info on this with Socket M? its not the easiest laptop to get to the CPU but ive done it couple times now

otherwise i was gonna ask if theres anyone/anywhere that can modify the BIOS to force it at full speed or prevent SpeedStep from lowering the multiplier when disabled
 
well im confident that turning Speedstep off lets me use the laptop on battery no problem and when its on it practically crashes instantly when running on battery. but when its enabled and not using battery, CPU-Z does show that multiplier does range throughout 6 to 14 no problem. now im not sure if Speedstep and CPU are the actual issue or not
 
talking to a guy on bios-mods forum, seems as tho its not gonna be a truly fixable issue. its may not necessarily be SpeedStep causing the issue. but i have made some progress to get around the issue. heres the thread
https://www.bios-mods.com/forum/Thread-Dell-E1505-need-help-with-SpeedStep-and-clock-multiplier

i can boot with Speedstep disabled in the BIOS, then use the program ThrottleStop to re-enable Speedstep and adjust the multiplier. without SpeedStep, changing the multiplier wont take effect. the good news is i can now reliably run at multiplier up to 12 (2.0ghz) stably on battery and 14 (2.3ghz) plugged in. the bad news is that anything above 12 on battery causes lockups, booting up will still be at only 1.0ghz. luckily being a pretty bare version of Windows XP with few startup programs and having an SSD the loading time is still decent, and a multiplier of 12 is still faster than the old cpu and is dual core instead of single.
 
the voltage is maxed at 1.25v and still crashes. but since its runs fine plugged in at 1.25v and 14 multiplier, i would think that's not the issue.

but i think what the guy there was getting as is the motherboard is not able to convert the battery voltage correctly at those voltages. when on battery i hear a buzzing noise when i set to 1.25v and it starts going away the lower i set it. when plugged in there is no noise. so he may be on to something. power might not be consistent enough when its set that high while on battery
 
I wonder what exactly the culprit is. I wonder if injecting a newer microcode would resolve the issues.
 
The new CPU has the same maximum voltage as the old CPU, just maybe drawing 8% extra amperage at maximum. I doubt Dell would've cut it so close on tolerances with the PSU or motherboard. Can you open the thing up and check for any dying/bulging capacitors on the motherboard?
 
Back