E4300 System Built

[Grimmy]

Thought I'd share a recent build I did for my Dad's B-day.

Case - $65 - AeroCool AeroEngineII-SSP Silver
MB - $113 - P6N - SLI
CPU - $133 - OEM E4300
VCard - $90 - Asus 7300GT
Ram - $110 - A-DATA 2GB DDR2667
HSF - $20 - Sunbeam CR-SW- 775

Edit:
DVD- $31 - LG Black 18X (for got to add)

Total cost without shipping: $562

PSU - 430W Antec - reused
250 gig HD - WD - reused
TV Capture Card

And for Temp Readings:

Thought some may be interested in the temps the budget PC I made for my Dad.

So these temps will be based off the clone (sunbeam) of the AS7 Pro. The room temps were around 76 F when I did these quick tests.

Also, I've tried speed fan, and MB5 software, and it looks like at this time, there is no chipset support that I can find for it. So basically it shows no temps, and I rely on MSI's app.

First off idle temps in windows:

Prime 95 - Core 0 load only:

Prime 95 - Both under load:

TAT - Core 0 load only:

TAT - Core 1 load only:

TAT - Both Cores on load:

So you can see the test differ (TAT vs Prime95) on the amount of load it does to the core(s).

I also decided to show pics of the PC case in the dark. It was kind of tricky to do, by increasing the time exposer, and trying to be real still. :lol:

Hope you enjoyed the lil amount of werk I shared on this thread.

 

[Grimmy]

Just a lil update:

Ran Prime95 over 6 hours

Few mins after cool down:

Super PI:

1.8 Ghz - 30sec
2.4 Ghz - 23sec

Not too shabby for a sub $600 upgrade for good old pops.

 

[skyguy]

Well done!! Good job bud.

 

[Grimmy]

Tanks.

Just wish I had the setup for myself. :lol:

 

[skyguy]

LOL, ain't it the truth.

 

[Gundam1s]

Nice job there man, but did you ever consider lowering the Vcore? Because I have a E4300 myself and runs below 1.3V @ 2.4Ghz.

 

[Grimmy]

I guess I overlooked one setting... <-smacks himself.

Doing some retesting. I guess I didn't have the ram voltage at 2.1 when I rechecked. Once I got that raised, I went back to stock vcore. And what ya know, its Prime95 happy. :lol:

So now the vcore at 1.296 (cpu-z). Temps reflected the change around 2 degrees cooler.

Thanks for the advice, I should have known better.

 

[akhilles]

Somebody stole my PC. nooesssssssssss!

I was on the fence about the 650i mobo choice. I looked at p6n sli. Everything is good except for the overclocking options. I looked at p5n-e sli. It overclocks great, but lacks proper cooling. Read through a thread of 30-something+ page. There's another with over 100+ page at anand I'm gonna read.

2 things about the p6n: odd placements of floppy & cpu power. My p5n-e's fdd is farther & with that case, the floppy cable can't be connected if a full-size pci card is installed.

Good o/c for a dad's pc. I reset my old c2d setup to default for my family pc. It's rather silent.

 

[Grimmy]

Heh, tanks for the OC comment.

About the Floppy drive, I didn't even connect one, and disabled it in the bios. Before buying the MB, I thought it was kinda weird placement, but I also thought it would still work out great to hook one up when I needed it (outside the case), which would be a bios update. And if my dad even need anything off floppy, he could use his laptop USB floppy drive to read off them.

Turns out later, when I was looking for bios update, it seemed that the MB had the latest on there. And one last thing, I ended up reading the instructions. MSI really does not want me to use the floppy to update the bios. They want you to use the floppy to get to DOS, but you have 2 choices from there.

1.) Using a boot up disk with loading a ram-drive into memory. So you copy the flash.exe and bios to memory (ram drive), then run it.

2.) Or.... having the flash/bios files on the HD in FAT32 format, and then run off the HD.

It seems theres a 3rd option, since the startup boot sequence can run of a USB drive.

So... I guess I won't be really needing the actual floppy drive hooked up at all. The midtower only has 1 3.5" bay, which I ended up using for a multi-flash card reader that came off the old system which hooked up to the USB header.

:lol: . o O (that was allot to type about the floppy)

 

[Grimmy]

Well, I thought I couldn't run Speed Fan, until I tried again with the latest setup installation. Seemed to take a real long time to startup. :lol:

I didn't take the time to make screen caps, but the readings for the cores (0-1) are about 10-15C cooler. So fresh out of my mind on idle temps:

TAT or CoreTemp .95 - idle temp 42C/44C

Speed Fan - idle temp 27C/29C

The other temps it picks up agrees with the MSI app that just reads a cpu temp and system temp.

Is the Speed fan reading a different sensor, like the older CoreTemp .94 beta? :?

 

[akhilles]

core temp has a .95 version. it's more accurate than speedfan, but it crashes at random.

speedfan isn't as accurate, at least it doesn't crash. the 1st time it runs takes a while. it has readings for other temps & fan speeds, etc.

 

[Grimmy]

Umm I'm aware of the 2 versions of CoreTemp. And the older version seems to have the same temp readings as Speed Fan.

I noticed if I ran a combination of sensor software (TAT/CoreTemp/MSI app), something crashes or locks up the system every now and then. Adding Speed Fan to run for comparing, I got abit weary. :lol:

Edit: I have CoreTemp .95 in the startup, so it's the only temp app I have setup to run. My dad hasn't complained of any crashes or lockups like I experienced running more the one sensor program.

Just wondered if its just the software values, or if its actually reading different sensors. I know there are 3, but I guess it would be silly if there were more then that.

So perhaps its just the software values which it determines the temp are different?

 

[akhilles]

I was reading an Intel forum: the Intel mod/dev locked the thread when a non-Intel dev asked for the core temp calculation. It might have to do with the inaccuracy of speedfan, core temp, etc. cuz the dev's may not know the formula.

If accuracy is your thing, use Intel's TAT. Google is your friend.

 

[Grimmy]

Hmmmm... I went back to the CoreTemp website, and took a look at the change log:

Core Temp Change Log

The one thing that caught my eye on the fixes was:

Version 0.95 - 17th March, 2007

- Fix: Revision F Turion/Turion X2 detected wrong.
- Fix: Core Temp fully loads one core on a dual core AMD CPU.
- Fix: Core Temp crashes on Dual Core Opteron 200 series systems.
- Fix: Core Temp misdetected certain Dual Core Opteron CPUs.
- Fix: Must set Core Temp to run as Administrator in Vista.
- Fix: Can't see Core Temp GUI if Core Temp was last closed minimized.
- Fix: C2D L2 rev. report very low temperatures. (Tjmax to 100C)
- Fix: Core Temp reported wrong temps after a while. (Driver conflict)
- Fix: Core Temp reported Max VID on Intel CPUs. (Now shows current)

Kinda wonder if that (Driver conflict) might be the case with speed fan. Accuracy would be nice, I mean why be mislead by, at the most, 15C? I just want to know what is more accurate. I seem to trust TAT more, but the thing about it is, it seemed to be more designed for the mobile processors just reading the help file, and trying to find more info on TAT, which is kinda hard to find. Even looking for an update version of it.

I saw one post on an E4300 reaching 90C before shutting down, so I'm guessing CoreTemp might be pretty close in determining the max Tjunction temp of 100c, though, not 100% sure, since I feel as though the max temp of 85c more realistic for desktops.

:lol: . o O (I guess as long as the PC works OC or not is what really matters)

Edit:

Finally got around (okay, I'm lazy) getting the CPU-Z info off:

http://valid.x86-secret.com/show_oc.php?id=199225

 

[Grimmy]

Just another update.

I was told of the CPU section that you can still OC with speed step enabled. At first, I thought no way, why would that OC guide for C2D have you disable it.

So, I enabled EIST and C1E, and ran Prime95 for 6 hours test.

Turns out it can be done:

You can see CPU-Z is reporting 2.4 ghz OC

When Prime95 is stopped, the CPU speed drops to 1.6ghz.

I found that kinda amazing, but what is really weird, it just jumps from 1.6ghz to 2.4ghz when the CPU goes into a load. That's quite a shift on ghz on C2D systems running at 3.2ghz though, namely the E6600. With it's speed step running, it still drops it to 1.8 I think. Still quite a hop from that to 3.2. I would think the system would freeze.

I was hoping to see a temp drop, but it seems the room it was in got abit warm (80-81F running prime95 for 6 hours) since the AC didn't kick on, or I should say the living room was at 78F, so the AC won't kick on. :lol:

Other then that, temps are basically similar if not the same even after a cool down for about 10 mins. I suppose I would have lower temps to show for it, if the room was at least 76F like I had done on the previous test above..

One last thing to mention, since I use a P3-kill-o-watt meter, there was a slight difference on idle consumption.

E4300 - with Speed Step off - Idle 113 watts
E4300 - with Speed Stop on - Idle 107-108 watts

E4300 - will not matter off/on - load @ 2.4ghz 160-162 watts

Anyhoo... thought it was interesting.