Building a "cheap" PC - Final ~$1100 (AMD945/4GB RAM/1TB/ATI 5770HD/550W/24" XL2370)

[Omnis]

Yeah, windows 7 is a piece of cake. I'm so glad apple had such an influence on microsoft that they had to make things simple.

 

[DQuaN]

2 of these in CrossFire would work nicely! http://overclockers.co.uk/showproduct.php?prodid=GX-237-SP&tool=3


I know, not cheap, but I just came across it and had to post it somewhere!

Looks like a really nice build. I'd hold off from overclocking for the moment. If you're new to this kind of thing I would take a while to enjoy it before you start the stress of messing with voltages and clock speeds etc.

 

[ROAD_DOGG33J]

Can we get some pics?

 

[icelt]

I would guess around 16500 before you start OCing. I might be low though. This is for 3dmark06 and not Vantage. I haven't run Vantage on my computer yet.

Right. Yeah I downloaded vantage and only chimed in with a paltry 9100 something. For a sec there I was wondering "what did I do?" I'll try 06 here in a sec. Or 1900 seconds. Damn 06 is a larger download than Vantage it looks like.

Edit....
Score came back as 15268. So all things considered I'd say you were pretty close, although I'd wish it were higher (not that I know any different at this point admittedly). But boy it seemed like it was bogging on the 2 CPU tests, didn't see the framerate go above 2 for either one of'em.

Yeah, windows 7 is a piece of cake. I'm so glad apple had such an influence on microsoft that they had to make things simple.

Easy pee-sy. I was surprised at how few ?'s I had to answer... maybe 10.

Looks like a really nice build. I'd hold off from overclocking for the moment. If you're new to this kind of thing I would take a while to enjoy it before you start the stress of messing with voltages and clock speeds etc.

Think I don't exactly have the best xfire board, it's a cheapo MB afterall . Yeah don't think I'll feel bold/desperate enough to start the OC fiddling for awhile.

Can we get some pics?

Sorry was AFK for a couple days there. I'll definitely put some up soon, however Omnis' build now has me wanting to clean up my cables a bit better. Not that I'll ever likely have any significant airflow issues, but dang his case allows for a CLEAN install.

 

[GT3mich]

Sorry was AFK for a couple days there. I'll definitely put some up soon, however Omnis' build now has me wanting to clean up my cables a bit better. Not that I'll ever likely have any significant airflow issues, but dang his case allows for a CLEAN install.

So does yours. I've built 3 computers with the Antec 300. Plenty of room to hide cables.

 

[Omnis]

Make sure you have long enough PSU cables. My ATX and CPU cables were a pain in the ass to get in their sockets.

 

[icelt]

So does yours. I've built 3 computers with the Antec 300. Plenty of room to hide cables.

True, but it doesn't have the huge back side access holes that Omnis' case has was basically my point.

Loosely related, I have an acquaintance who wants me to build an exact copy of my machine for him. What is the reasonable going rate for building CPU's these days?

 

[ROAD_DOGG33J]

Loosely related, I have an acquaintance who wants me to build an exact copy of my machine for him. What is the reasonable going rate for building CPU's these days?

It's not exactly rocket science [if you didn't actually mean CPU], so I would say maybe $30-40. You could go lower depending on your kindness or higher if they're loaded.

 

[icelt]

It's not exactly rocket science [if you didn't actually mean CPU], so I would say maybe $30-40. You could go lower depending on your kindness or higher if they're loaded.

Even rocket science is not rocket science, but I know what you mean. Yeah I should have asked more along the $/hr going rates, as I will be installing the OS, tweaking the monitor, installing updates and downloading basic programs and such. Not overly challenging stuff to be sure, but it all adds up and charging a reasonable rate for labor is perfectly fair.

Anyway de-nesting the cables will post pics after.

 

[icelt]

Now for the pics (no idea why the quality is so meh, but I'm no great photographer either).

Monitor and speakers:

Access side prior to cable cleanup:

Close up 1 PtCC:

Snakes alive.


Close up 2 PtCC:

Snakes. Snakes I tell ya.


Close up after cable cleanup:

Notes:
* See how I craftily dealt with the lower left excess power lead. Didn't want that thing falling into the PS fan intake .
* After reading through the MB instructions I learned that in fact for PC 1600 RAM the secondary set of slots (white ones) are recommended over the primary (blue ones). I read that section at least 3 times and it always said the same thing so I decided to trust it .
* I like how I employed the latch of the CPU fan/heat exchanger as a routing means for the fan PS wires.
* Obviously the MB power cables are just too short to route any differently, so I wrapped them together as best I could to minimize ungainliness.


Back side ACC:

Note how there is really not much to note.


Front no flash:

Oh... blue & red lights. Mystical.


Front with flash:

 

[Omnis]

Oh my god, you weren't kidding about the wires.

 

[ROAD_DOGG33J]

How close is the case to the wall?

 

[icelt]

10-12" I'd say.

 

[GT3mich]

Here is my wife's computer.

You're cables should be long enough to hide them under the MB. Just. The only thing is you'll have to take it apart again.

 

[icelt]

You're cables should be long enough to hide them under the MB. Just. The only thing is you'll have to take it apart again.

Ohh, you were just waiting to spring that on me weren't ya . J/K . I'll think about it, basically a complete disassemble.

 

[Pako]

Ohh, you were just waiting to spring that on me weren't ya . J/K . I'll think about it, basically a complete disassemble.

I agree on the wiring. It's not just for looks, but it also promotes better air flow for your system as a whole, which in turn provides cooler running components which is a good thing. Look at the air flow that is being restricted right above the CPU fan. This is a no, no..... Off to the side is a maybe, but definitely not directly over it. You would have to run some load temps to see if it actually makes a difference, but it's something to consider.

I will say it does look good though, especially for your first build.

 

[ROAD_DOGG33J]

Look at the air flow that is being restricted right above the CPU fan. This is a no, no..... Off to the side is a maybe, but definitely not directly over it.

Yeah...if you don't want to rewire everything you should at least consider getting the wires off of the CPU fan. It should be fine either way, but this is a pretty simple thing to do.

 

[icelt]

Yeah I know you guys are speaking earnestly. Although I don't necessary agree that the power cables above the CPU create a significantly adverse airflow restriction. Think of when you stand in front of a open door with a good breeze going through it. Sure you are restricting the free flow path by some amount, but it still is quite breezy . Anyway, I fully agree that regardless of amount it would be an improvement to re-route the power cables as suggested. Just a pain in the 🤬 lol.

On a related note I ran the Prime95 stress test (thanks to Omnis' thread ) for a couple of hours and saw a peak CPU core high temp of 62 degC. This was with the current assembly conditions, both case panels attached and the 2 case fans set to low speed.

Now I have some follow up questions:
1 - AMD's website states a max recommended operating temp for my CPU of 62 degC. Now what's the communities take on this? 62 too high? 62 too low? I know cooler is always better on the MTBF curve, but that's not to say the 200 hr difference (just throwing out a number) between operating at 62 and 70 degC is worth the concern.

2 - I was really hoping as I read through the MB manual that I'd be able to take advantage of its built in variable fan speed control. Alas I didn't see the 3 prong signal adapter on the PS's fan lead that I think is required to make it work. Anyone know if my PS is compatible with the MB's VFS control? If so, do I need to buy a special lead/cable to make it work?

 

[ROAD_DOGG33J]

2 - I was really hoping as I read through the MB manual that I'd be able to take advantage of its built in variable fan speed control. Alas I didn't see the 3 prong signal adapter on the PS's fan lead that I think is required to make it work. Anyone know if my PS is compatible with the MB's VFS control? If so, do I need to buy a special lead/cable to make it work?

That fan should be plugged into the motherboard.

 

[icelt]

That fan should be plugged into the motherboard.

Well both fans have a single 4-prong male/female power connector and a single 3 speed fan switch. That's it. The way the MB manual describes it, I'm led to believe that if the PS is compatible I'd plug a fan control wire attached to the fan power lead into the MB. I'd then set both fans to high speed and then the MB would direct the PS, based on a temp setpoint, to vary the fan output voltage to effectively change the fan speed. As near as I can tell this PS is not compatible, but I am hopeful that I'm wrong .

 

[Pako]

Now I have some follow up questions:
1 - AMD's website states a max recommended operating temp for my CPU of 62 degC. Now what's the communities take on this? 62 too high? 62 too low? I know cooler is always better on the MTBF curve, but that's not to say the 200 hr difference (just throwing out a number) between operating at 62 and 70 degC is worth the concern.

Re route the Mobo and +4v lead to the side of the CPU and get those temps down to 61 deg C.

I would consider this as a routing option. Run the +4v lead under the graphics card and along the red path. Run the Mobo lead off to the side of the mobo, using zipties to keep it in place along the green path, preferably under the rest of the horizontal wiring. This would only take about 5 minutes and would not require you to remove the mobo, although you would have to take the graphics card back out to seat the +4v lead. Just a thought. Either way, it's TONS better than the original spaghetti mess you had going on in the earlier pics.

Since you have some stress test temps, I would be curious if it made a difference at all...., like I said, just general rules is all that I am speaking of.

 

[ROAD_DOGG33J]

Well both fans have a single 4-prong male/female power connector and a single 3 speed fan switch. That's it. The way the MB manual describes it, I'm led to believe that if the PS is compatible I'd plug a fan control wire attached to the fan power lead into the MB. I'd then set both fans to high speed and then the MB would direct the PS, based on a temp setpoint, to vary the fan output voltage to effectively change the fan speed. As near as I can tell this PS is not compatible, but I am hopeful that I'm wrong .

Oops screwed up there. I thought you were talking about the cpu fan connector. Does that one vary it's speed? There should be a setting in the BIOS.

I forgot the story with the system fans, but I think if they are connected through a 3-pin connector into the motherboard then the variable fan speed might work.

In either case, they would only be connected to the motherboard.

edit: What page of the manual is this on?

 

[icelt]

Re route the Mobo and +4v lead to the side of the CPU and get those temps down to 61 deg C.

I would consider this as a routing option. Run the +4v lead under the graphics card and along the red path. Run the Mobo lead off to the side of the mobo, using zipties to keep it in place along the green path, preferably under the rest of the horizontal wiring. This would only take about 5 minutes and would not require you to remove the mobo, although you would have to take the graphics card back out to seat the +4v lead. Just a thought. Either way, it's TONS better than the original spaghetti mess you had going on in the earlier pics.

Since you have some stress test temps, I would be curious if it made a difference at all...., like I said, just general rules is all that I am speaking of.

Damn you! Now in the name of thermodynamics I have to re-route and retest . I bet I don't see more than a 0.5 degC improvement . We'll see though...

Oops screwed up there. I thought you were talking about the cpu fan connector. Does that one vary it's speed? There should be a setting in the BIOS.

I forgot the story with the system fans, but I think if they are connected through a 3-pin connector into the motherboard then the variable fan speed might work.

In either case, they would only be connected to the motherboard.

edit: What page of the manual is this on?

Yes the CPU fan speed varies just fine.

My screw up this time. I actually found that information in the Antec manual, pages 4-5. The PS does have a 4-pin connector on the fan power lead, but not a 3-pin like you mentioned and the MB manual states as being available for fan connection.

 

[nick09]

You can control the three pin fans by using a fan controller. They have to be a four pin fan just like the cpu fan to be controlled manually by the motherboard. This is a good example of a fan controller. You can control about 20 watts per channel(so you can daisy chain multiple fans on one channel. That's as long as don't you don't exceed the 20Watt per channel limit). I still have the silver model and it's a good product.

 

[Pako]

Damn you! Now in the name of thermodynamics I have to re-route and retest . I bet I don't see more than a 0.5 degC improvement . We'll see though...

YES! In the name of thermodynamics, you must! :tup:

 

[icelt]

So whomever said the power cables re-route would only take like 5 mins was absolutely correct. Also since there isn't enough clearance between the MB and back plane there'd be no use/benefit try and route them beneath anyway.

Final Routing Pic:

Now in the name of thermo-damn-namics... WE TEST!!

 

[Pako]

So whomever said the power cables re-route would only take like 5 mins was absolutely correct. Also since there isn't enough clearance between the MB and back plane there'd be no use/benefit try and route them beneath anyway.

Now in the name of thermo-damn-namics... WE TEST!!

Looks just like a picture I saw once... Now lets see those test temps! 👍

 

[ROAD_DOGG33J]

I like the routing now much better. It looks a good bit cleaner.

 

[Pako]

I like the routing now much better. It looks a good bit cleaner.

Yes, I agree!

 

[icelt]

Looks just like a picture I saw once... Now lets see those test temps! 👍

Yez, the final result is eerly similar to your sketch . Recalled the statement just not the stator. Could of looked, but I was feeling lazy.

So far the peak core temp post re-route has been 60 degC, however the ambient temp in my condo has been a good 3 degF lower as well. Sooo... to do it right I'll probably have to re-run the tests with both routes in a very short time window to help minimize the influence of "steady state" conditions.

I like the routing now much better. It looks a good bit cleaner.

Thanks, you guys were right. 👍