Page 2 of 2 FirstFirst 12
Results 17 to 24 of 24

Thread: How to silence a noisy Dell?

  1. #17
    Senior Member
    Join Date
    Aug 2003
    Location
    Richmond
    Posts
    217
    Thanks
    0
    Thanked
    0 times in 0 posts

    BIG OCs [was: How to silence a noisy Dell?]

    chrisf6969,

    Well, had a good muck about...
    • Set it at 3:2
    • Put the FSB up from 252 to 260 (not a huge increase)
    • Upgraded the BIOS to the new ASUS #12 for the P4P800 (was at #8)
    • Gradually upped the Vcore from 1.675 right up to 1.875
    • Set Vdimm from 2.75 to 2.85 (don't have any more in the BIOS, but as the mem is running well below rated shouldn't need to up this anyway)

    And every time it freezes after about 2 mins.

    HOWEVER I found that it's SETI that makes it freeze - it's in my Startup group. When I killed it quickly I actually got to run some Sandra benchmarks (CPU slightly better, memory much worse as expected). Turn on SETI and it freezes which is a bummer as it's a SETI crunch box!

    The CPU temp doesn't cross 42deg so that's not the prob here.

    So seeing as you can get a 50% OC out of your 2.4 and [b]I can only get a stable 25% out of my 2.6[b], what gives??!!

    Is there a difference between these CPUs and the rest of the line right up to 3.2Ghz, other than the multiplier?
    Why do some CPUs OC better than others?
    Is it simply that some plants fab them better than others?
    If you can get 50% OC does this mean we'll see go-l.com and others shipping 4.6Ghz VapoChill machines???

    Presumably there's a physical upper limit to the max speed of these chips and it's been hit at about 4.1Ghz on the nVentiv demo machine? http://www.nventiv.com/index.php?pageid=44

  2. #18
    Banned
    Join Date
    Jul 2003
    Location
    House without a red door in Birmingham
    Posts
    1,595
    Thanks
    0
    Thanked
    0 times in 0 posts
    Could simply be the limit of your particular processor, even 2 CPUs off the same line (even one straight after the other) will often have different max o/c's. Anyway it is best to take gradual increases rather than jumping from 252 to 260 (3.28ghz to 3.38ghz) just beacause you upped the voltage on RAM & CPU as well as using 3:2 ratio. You may not get much higher anyway but it's best to keep normal voltages and up the FSB 1-2mhz at a time (rem it's QDR so 8mhz is really 32mhz extra FSB). Only gradually up the voltage of the CPU when it seems to max out, same for the RAM and at 3:2 you'd need 300FSB (QDR=1200FSB) to hit the normal speed for your PC4000 so no extra voltage is needed (and can cause instability or damage).

    There could still be a chance of you getting higher by tweaking BIOS settings, esp switching off HT as some find they can get a good 0.3ghz extra speed with HT off (ie 3.3ghz to 3.6ghz). For general Windows usage you are best to keep HT but for gaming pure ghz is best. There is a chance that 3.4ghz actually o/c better than 3.6ghz CPUs mostly because the 3.6ghz kept a high price for a long time so the 3.4ghz ones had a higher turn over and hence you had a better chance of getting the latest revision and best o/c'er. At the same time the multiplier of the 3.6ghz should mean FSB (etc) become less limiting in the o/c so it should balance out. Bottom line 3.3ghz is a great o/c and faster than the current top P4 (esp with increased FSB) so even if you don't get anything more it's still a great boost!

  3. #19
    Senior Member
    Join Date
    Aug 2003
    Location
    Richmond
    Posts
    217
    Thanks
    0
    Thanked
    0 times in 0 posts
    Hi Austin,

    I jumped from 252 to 260 as I figured I would need really big FSB increase (bigger than 8Mhz, but 260 is a nice round number) to get to match the performance of 252@1:1 when running at 3:2.

    ie there's no point running at say 258@3:2 when it'll do 252@1:1, no?!

    I agree though, even though I've been left wanting more, it's better than the current top of the line and much cheaper too!

  4. #20
    Senior Member
    Join Date
    Jul 2003
    Location
    New Orleans - baby!
    Posts
    340
    Thanks
    0
    Thanked
    0 times in 0 posts
    What timings do you run your ram at each of the following settings? And how are your benchmarks for each setting?

    252 with 1:1 - ram at 504Mhz - (guessing 3-4-4-7 & 5500Mbs)
    260 with 3:2 - ram at 347Mhz - (guessing 2-2-2-5 & 4900Mbs)
    260 with 5:4 - ram at 416Mhz - (guessing 2-3-3-7 & 5300Mbs)

    I get
    300 with 3:2 - ram at 400Mhz (2-3-2-6 - 5800Mbs)
    Last edited by chrisf6969; 17-11-2003 at 08:03 AM.

  5. #21
    Senior Member
    Join Date
    Aug 2003
    Location
    Richmond
    Posts
    217
    Thanks
    0
    Thanked
    0 times in 0 posts
    The memory is Adata PC4000 CL3-4-4-8 which I haven't previously been able to run at anything other than by SPD.

    At 3:2 the mem bandwidth was way off the 5500 I get @ 1:1, 4500 I think.

  6. #22
    Senior Member
    Join Date
    Jul 2003
    Location
    New Orleans - baby!
    Posts
    340
    Thanks
    0
    Thanked
    0 times in 0 posts
    what about at 5:4?

  7. #23
    Senior Member
    Join Date
    Sep 2003
    Posts
    254
    Thanks
    0
    Thanked
    0 times in 0 posts
    Originally posted by shredisn@tdead
    The memory is Adata PC4000 CL3-4-4-8 which I haven't previously been able to run at anything other than by SPD.

    At 3:2 the mem bandwidth was way off the 5500 I get @ 1:1, 4500 I think.
    Excuse the noob but whats SPD mean?

  8. #24
    Senior Member
    Join Date
    Aug 2003
    Location
    Richmond
    Posts
    217
    Thanks
    0
    Thanked
    0 times in 0 posts
    Didn't try 5:4 this time (had a while back without any luck). Figured that as I couldn't get any significant increase at 3:2 then I wouldn't bother trying 5:4 again

    Have you tried running at 5:4? Assuming your ram is rated for it, that is? Fast memory that'll run that tight would be expensive though. It'd sure generate some impressive benchmarks in your system though.

    sparky191,
    If I remember right SPD = Serial Presence Detect, and "by SPD" simply means that the mobo uses the "preferred"/default timings that are programmed into the ram chips by the manufacturer.

Page 2 of 2 FirstFirst 12

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •