View Full Version : cel 1.2 Philippines testing
hi,
I had cel 1.2 Malayan chip before and it didn't do 1600 stable enough. (it needed 1.75V to even pass Prime wo/ errors a short time, and over 1.8V and window open cooling to reach 1600...)
Well, now I got a new chip, after selling the old one, Philippine cele 1.2 OEM. (SL5XS, 7140A302-0121). It ran Prime 30 minutes (1,65V, 1603,63Mhz), but then Prime halted on error. Now I'm testing it w/ 1.70V w/ Vmodified Powerleap adaptor, computer case closed... though I got improved case cooling now.
It looks like this chip is anyway much more overclocking friendly than the last chip, cause it doesn't need as much voltage as the last chip. I'll post more results and info later.
dOER
This is what you get with different chips. Even though they are essentially the same, no chip works identical.
Looking forward for the results :)
What kind of cooling are you using?
I've seen people go 1.7 on that chip with 1.825 with good cooling of course.
Hope you hit more
Yodums
I have a Malaysian chip and it's a ripper,running with standard HSF at 1608MHz @ 1.77 volts.It runs a bit hot - approx 40 degrees idle 47 degrees at full load.Not real hot I suppose,but i'd like it cooler.Does anyone know a safe temp and voltage for these chips yet?
Originally posted by Whitey
I have a Malaysian chip and it's a ripper,running with standard HSF at 1608MHz @ 1.77 volts.It runs a bit hot - approx 40 degrees idle 47 degrees at full load.Not real hot I suppose,but i'd like it cooler.Does anyone know a safe temp and voltage for these chips yet?
Standard HSF?!
Did you change the thermal paste and lap it or something?
Yodums
Running mine @ 1608, 1.62 v.....temps are 28-32c idle, 34-35c load.....very happy with mine. Gonna go higher as soon as my better heat sink arrives....
Haven't lapped it yet,but I did apply some arctic silver 2. OC2K,I wish mine ran that cool,and at that voltage.Is there a way you can accurately measure the cpu vcore with a multimeter??,I heard the Asus boards ,with Asus Probe software,over- read the voltage and temp.Also ,is Sandra 2002 burn-in wizard a good stress test?I found CPU Multimedia Test gives it a fair workout so I just loop this for 100 times at 100% CPU usage.Is this a good enough test for stability?
I found CPU Multimedia Test gives it a fair workout so I just loop this for 100 times at 100% CPU usage.Is this a good enough test for stability?
Well, during stability testing, I have found that looping the demos for "Serious Sam" is a good test. I discovered that looping the SS demos will show instability, with my system, long before prime95 or Looping 3dmark2001 will. Recently, while testing my 1.0a for max OC, I ran prime95 and looped 3dmark for an hour with no problems. Then I closed everything and ran the Seriuos Sam loops, and it locked up within 5 seconds into the opening fly-by. After dropping the fsb down 2 mhz, it would loop SS forever, as well as do anything else stable. So instead waiting all night to see if a test failed or not, Serious Sam seemed to give me quicker results. If you have the game, give it a try. If not, then get the Serious Sam version 1 demo, and go to town.
BTW, so far the most I managed to squeeze out of my 1.0a is 1420 @ 1.7v, with cheap air cooling:)
Sklathill
01-17-02, 12:30 AM
ARGH! Why do I have to get an OEM Cel 1.2 Philippines that can't hit 1.53? DOH DOH! :) (When I first got it, it wouldn't even POST at 1.6, even with a vid pin mod...) As for stability testing, I do Prime95 + Q3A and CPUBurn. I've never seen anything that gets my temperatures as high as cpuburn...
Whitey....You woulden't happen to be off the OCAU forums would ya?
Rooski,yes I have that game,do you mean the demos that are on at the start or do you have to add a command line in the console?What are your temps at @1420 ??
G'day Flu!d,it's a small world isn't it.Have you cranked your cpu up further yet,or are you playing it safe?
new results...
it ran overnight Prime w/ no problems (1603Mhz, 1.70V). I think about 11-13 hours :) until one family member accidentally shut down the comp. :O And I had to rebuild my raid-1 after that :/ Dunno if the rebuild was due to illegal shutdown or overclocking failure.
Áfter raid rebuild I tried to get 1684+ Mhz, but @1.70V it wouldn't go further to windows than "press f8". @1.75 & 1.80V it gave me blue screen before loading windows. I set my memory pciclock - 33Mhz. My last celeron couldn't get to windows @ 1684+ either. So it could be my motherboard, psu or the cpu...
Now I started Prime all over @1603Mhz (1.725V)
Cooling is pretty heavy aircooling with good case and cpu cooling... cpu cooler is of copper and the fan is 37cfm, 7000rpm. I'll write more after seeing if this Primetest passes or fails.
*
Sklathill, be sure your wiretrick is working... my last celeron posted 1600Mhz, 1.60V and it was Malay.
dOER
Rooski,yes I have that game,do you mean the demos that are on at the start or do you have to add a command line in the console?What are your temps at @1420 ??
I just launch the game and don't do anything. The fly-by loads first, followed by the demos. They will loop indefinitely, until you hit escape to get to menu.
As for my temps, I'm using a decent, cheapy Antec
cooler I got at Best Buy for $15. It works better then the stock Intel cooler. I'm running the fan on 7 volts to keep the noise down. The difference in cpu temp is just a few degrees between 12v and 7v. I chose less noise over less heat. Using Hardware Doctor on my ST6R, my cpu temp, under load, hovers at or just below 50c. This is always about 2 or 3 degrees above my case temp (closed case). I'm not concerned about this temp, at all. Intel states, in their white paper, to maintain a case temp of below 69c, for this cpu.
I'm also using an Antec 350w PSU (dual fan). I thought it was to noisy, so I yanked out the bottom fan, and have just the rear fan operating. After doing this, my case temp is the same as is was with both fans running. I do not use any additional case fans.
I'm not a purist overclocker, but an econimist overclocker. I like to get the most "bang for buck". I also refuse to increase fan noise, just to gain a tiny bit of speed that will only be noticable in benchmarks.
BTW, I discovered an instability at what I thought my top speed would be, so I'm back down to a 139 bus (1407 according to CPUinfo). Overall, I'm really happy with this cpu!. I may follow the same upgrade path for my secondary rig :)
Running mine @ 1608, 1.62 v.....temps are 28-32c idle, 34-35c load.....very happy with mine. Gonna go higher as soon as my better heat sink arrives....
What are you using to cool the chip. My temps are at 42c full load at 1.703ghz with the coolermaster and 7000 rpm delta on there.
br0adband
01-17-02, 10:04 AM
Extol:
I have the same mobo and the same chip, I'm curious as to how you're getting 1.6V for the vcore since the voltages in the BIOS don't support that particular setting.
Since the 1010 BIOS was released, the Tualatins do 1.475, 1.525, 1.575, 1.625 and 1.675. Did you do the voltage mod to the CPU?
Just curious as I'm still looking to squeeze out a bit more from this CPU.
br0adband
Sklathill
01-17-02, 02:58 PM
Originally posted by doer
Sklathill, be sure your wiretrick is working... my last celeron posted 1600Mhz, 1.60V and it was Malay.
My wiretrick is definitely working. MBM is telling me I'm at 1.72 v (set to 1.725 in the bios).
This just isn't the great chip I hoped for. Actually, just a few minutes ago, Prime95 gave me a rounding error about half an hour into the test...1.52 GHz @1.75 v is just enough for Prime95...might crash in something more demanding on the CPU at that voltage, so I guess I'd have to be running at 1.8 v for best stability. Up the FSB to 1.53 (127 FSB in bios, though it's almost a MHz faster judging by SiSoft) and even at 1.8 v, Prime95 will not finish. I barely post at 1.6 as it is. 1.6 @ 1.7 v will not even post. 1.6 @ 1.8 will post, but will give errors on Windows startup.
It's not a matter of cooling either. Though my sig says running the Glaciator at 3300 RPM, I've also run it full out as well as a bevy of other heatsinks. (Zalman 3100, Delta-Thermoengine, Dr. Thermal, Delta-Mini Copper Orb) The Glaciator full speed does the best of all of them, but will not make the chip stable at 127 FSB. Glaciator full speed keeps the chip below 45 when CPUBurning and below 40 when using any other program, be it Prime95, Quake, 3DMark, or SiSoft.
Simply put, OEM Philippines Celeron 1.2 will not guarantee 1.6. Despite one's best efforts, overclocking still relies heavily on the luck of the draw...
Rooski,I ran the demo for 3 hours,no worries at all.I also did the benchmarking test and set the graphics to the "extreme quality"setting,for 1 hour.This really works your system,but i still had no problems at all.It's passed every test i can throw at it,played heaps of the latest games on max settings,so for now I will stay it is stable.I'm happy as a pig in **** with this cpu.I have 2 voltage steps to go,but i'm not going to push it at higher MHz,don't want to kill a sweet cpu like this.
Sklathil,just out of curiousity,which pins did you wire,vid 1 to vid 2 or vid 2 to vss?When I wired vid 2 to vss on my cpu ,the bios and mbm told me i was getting voltage adjustment,Asus probe and Sandra told me otherwise.Regardless of what i set the voltage to,I was only getting the new default-1.65v.Then I wired vid1 to vid 2 and the problem was solved.Also,I wrapped the pins with very fine speaker wire(one strand)not the "U" method in the socket.IMHO this gives a more reliable connection.
Sklathill
01-17-02, 05:00 PM
I wrapped a single strand of testing cable around vid 1 and 2. There's no question that I'm getting the proper amount of voltage, none at all. This just isn't as good a chip as others have. Heck, it won't even post at 1.6 volts for 1.44 GHz...
There is one thing that I could try out, and that's running a different PSU in here. As I ramp up the voltage, my 5 volt rail does a pretty unhealthy dip into the 4.9s and 4.8s. However, I tried running at 1.6 @ 1.8 v, except this time with three of my hard drives and my two optical drives disconnected from the PSU. On bootup, it got stuck at the Windows Registry Checker. :) Not fun.
Right now I'm popping down my voltage to see what is the least amount needed for stability at 1.34 GHz. Testing at 1.425 v right now...
I am using YS Tech 40cfm fan on AlphaPal 6035 with 2 80mm case intake fans and 1 120mm out take fan... but I am going higher as soon as my new MCXC370 Swiftech HS arrives....one of the best out there.........I will report what happens.......I am happy as a peach with this setup.....:)
I broke down and ordered an OEM Celly-T 1.2GHz today from Augustus Technology, Inc (http://www.augustustech.com/) for $82.00 to throw into my Abit ST6. We'll see what kinda luck I get -- if any!;)
Rooski,I ran the demo for 3 hours,no worries at all.I also did the benchmarking test and set the graphics to the "extreme quality"setting,for 1 hour.This really works your system,but i still had no problems at all.It's passed every test i can throw at it,played heaps of the latest games on max settings,so for now I will stay it is stable.I'm happy as a pig in **** with this cpu.I have 2 voltage steps to go,but i'm not going to push it at higher MHz,don't want to kill a sweet cpu like this.
Yeah, Whitey! Nice chip you got there! Congrats on the luck of the draw! If I had a good one like yours, I'd be nervous about pushing it, too. I'd probably just keep the vcore as is, and up the bus 1 mhz at a time, and use it for awhile to see. I wouldn't try any big steps at this point.
Hey Whitey! A fellow OCAUer!
It 'is' a small world!
I'm just playing it safe with my chip ATM, although I know it will pull the same speeds as yours at pretty much the same voltages as i've tried it before....Hell, we both bought our chips from the same vendor at pretty much the same time! (probably in the same week!)
I'm very comfortable where I am ATM, voltages are nice and low and temps are spot on....Not to mention peformance! I guess i'm not a 'hardcore' overclocker....Just the tinkering and thinking type!
Good to see ya here!
:D
GOOD news: now I've ran @ 1603Mhz, 1.825V (case closed) over 25 hours Prime95 and it seems to be stable. I ran some games too and I couldn't get them crash either (AOK TC) and tested that Serious Sam demo.
BAD news: high voltage... still @ 1.775 would crash AOK TC. It exited to Windows. BUT Prime kept running wo/ errors even after the game crashed :O Hmm. Wierd.......?
And I also updated my PSU to 300W after I had to about 5 times to rebuild my raid-1 array. I didn't lost any data. This same rebuild-thing happened to me w/ the Malay celeron too. After replacing PSU I haven't been able to reproduce rebuild...
I can't still boot at 1684+Mhz to Windows even @1.825V. It gives me blue screen w/ "Driver IRQL not less or equal". I don't know if it's my memory. I'm not sure if the 140- pci bus memory speed setting works.
more specs:
Mobo: Abit vt6x4, slot-1 + Voltage modified Powerleap T -adapter (cpu voltage cannot be read by any proggram)
Cpu:Philippine cele 1.2 OEM, (SL5XS, 7140A302-0121)
Mem: Spectek 256Mb, pc133
Case temperatures: +22*C down, +24 - +27 *C up
Room-temperature: +20-23 *C up, down +18 - +20 *C
Case-cooling: rear down 1x8, 1x9.2 in, back down 1x6 out, back up 1x8 in (w/ airduct to near cpu-cooler intake, I don't know if it actually helps or makes cooling worse by taking airspace), case top 2x 9.2 out, rear up 2x4 out
Cpu-cooler: copper, GlobalWin CAK-II 38 (w/ some Titan silver stuff and leftovers of Arctic silver)
PSU I've taken out of the case, so the air can move better inside the case.
Other: geforce 2mx, 2x lan cards, TV-card, ide raid-controller, 2x ide hdd, 1x cd-rom....
funny... I cooled my room down (+10-12*C), kinda easy when it was -12*C outside :D and could run my cele 1680Mhz. Normal room temperature (+18-22*C) it still locks loading Windows 1680Mhz.
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.