Advice on new build !

[mfenn]

Read my link. If you don't know about Bitcoin, though, look that up first, for some background. AMD can't just turn up supply on a dime, so this is what happens with increasing demand.

And if they do increase supply too quickly and demand suddenly falls through the floor (not unlikely given cryptocoin volatility), they are in the horrible position of sitting on a bunch of unsold inventory.

P.S. +1 on the XFX, after seeing the prices.

+2

 

[kokorochin]

Hello guys,

I just wanted to thank every person that took the time to help and advise me. Just know I really appreciate it. I took each and every advice into conisideration, as well as spending countless hours reading dozens and dozens of reviews for each component. In the end, I made my own choices and I am proud to announce that I have ordered all the parts. I went a bit over budget, but heck.

For those interested, here is the final list of the parts :

Board : MSI Z87-G43
CPU : Intel Core i5-4670K
GPU : Sapphire R9 280X Toxic Edition OC 3GB
RAM : Gskill RipjawsX F3-14900CL8D-8GBXM
SSD : Samsung 840 Evo 120gb
Fan : Cooler Master Hyper 212 Evo
PSU : Cooler Master V550S
CASE : 500R
Dvd : LG GH24NSB0
HDD : WD Blue WD10EZEX 1TB

Once again, big thanks to everyone involved in my decision process.

Take care, and see you around 🙂

 

[Sleepingforest]

Maybe not the ideal PSU or case, but I guess comes down to what you're willing to spend. I hope you enjoy building and using your new PC!

 

[kokorochin]

What's wrong with PSU and case ?

I have the opportunity to get a used 500R for half the usual price. Also, the 500R have received very good reviews.

The PSU is what I spent the most time researching, along with the GPU, I'm confident ! 🙂

 

[Sleepingforest]

See, this is why I need to read threads through very carefully before opening my mouth (keyboard?). Getting a used 500R half off is certainly a great deal. There's nothing wrong with the PSU, but you could have saved a bit of money and gotten a slightly more reputable unit in the XFX 550W. That's all.

 

[kokorochin]

I'm considering swapping the 500R + Z87 G43 for either a 350D or Fractal Arc Mini R2 + Z87 G43 mATX.

What do you think ? I would be more expensive, but take less space, which would be nice.

Looks like R280X Toxic + 212 EVO will be able to fit.

Worried about the R280X TOXIC obstructing the RAM slots ?

 

[mfenn]

The R9 280X Toxic shouldn't collide the DIMM slots on the Z87M-G43, but it will probably make operating the latches on the bottom of the DIMM slots a total pain. You will likely have to remove the GPU before changing anything about the RAM.

Regarding the case, both the Arc Midi and 350D are really nice. I would personally take the 350D because I don't like having the buttons on the top of the case, but that's really up to you.

 

[kokorochin]

Hey guys, I'm back, with bad news... I finally got all the parts and was able to build my long awaited computer...


So I bought a R9 280X Toxic two weeks ago. It got tremendous reviews everywhere so I decided to break the bank and get it despite my low budget. So here's the problem :

Out of the box, no overclocking no nothing, had artifacts, flickering, black shapes, crashes, all both in 3D applications as well as in 2D desktop/web browsing.
I tried many different drivers, tried to underclock, tried to up the voltage. No success. After 10** hours of trying various things, I sent the card back. RMA.

Today I just received the replacement card, and exactly the same issues are happening. Tried many drivers but no change (including the Catalyst 13.30). As soon as I start Furmark, the display goes out of control after a while, and if it doesn't, the application simply crashes after a while. In 3DMark and games, same things. Some games even crash 2 minutes in. Actually web browsing and desktop are fine, before you start any 3D application. But after display becomes corrupted in 3D applications or benchmarks, then 2D will be damaged and corrupted until the computer is rebooted.

(Card never goes over 80°c - everything is normal on that department)

I sent a ticket to Sapphire, but honestly I don't know what to do. I'm really sad and angry. My build is brand new and I haven't even been able to enjoy it yet. 2 weeks since I got the part, already one card sent back... Now this...

What should I do, send back the card and go for another brand ? I doubt Sapphire will magically find a solution to this problem.
Some are saying it's the PSU, other that it's the motherboard...

My specs :

500R
i5 4670K
MSI Z87-G43
CM V550S
Ripjaws 14900 8Gb
Samsung 840 Evo
Windows 7 - 64 Bits
Core stocks on everything

 

[circusslaughter]

http://www.logicalincrements.com/

 

[mfenn]

What happens if you remove the GPU and run Furmark on just the IGP? The performance is going to be laughably bad, but it will tell you whether or not the rest of your system is OK.

If you don't get any errors, then the problem is one of two things:

(a) you got another bad card (did you check the serial number to make sure they didn't send you the same card back?)

(b) the PSU isn't up to the task of powering the overclocked R9 280X. You can test this by hooking up a multimeter to the +12V output of your PSU and watching the voltage under load. If if sags more than 5% (11.4V) I would start to suspect the PSU. You can also measure voltages in software, but I don't trust those.

 

[kokorochin]

(Tried with a bad 800W yesterday for a short while, same problems)

Just got my hands on an Antec 1000W from another computer shop today.

Tomb Raider fine, Skyrim fine, 3DMark fine.

Furmark worked for 5min, and temps were 5°c lower than with the other PSU's. But 5minutes, now artifacts are everywhere. Wow, it's crazy ! Furmark crashed and the artifacts are out of control, way worse than anything else I've seen so far.

 

[kokorochin]

1. How come when using the Antec 1000W, the graphic card stayed between 65 and 75°c, as opposed to when using the Cooler Master V550S it's at 80°c in full load ?

2. Both shops I've been to, as well as the computer technician from the company that sold me the card ; they ALL tell me it's my 550W PSU that isn't strong enough. They say that Sapphire recommends a 750W and hence they might now accept taking back my card as they say I may have damaged it with such a low PSU.

3. On different reviews I see that the power consumption of the R9 280X Toxic in full load ranges from approximately 350W (games) to 469W (in benchmarks).
What if I got a model that requires 430-450ish Watts in full load ?
(those numbers vary greatly from one review to the other I have simply taken both extremes, Kitguru says 323 under furmark and 236 in gaming for example ; tweaktown 155/446 and overclockersclub 117/351)

Are you really, really, really, super sure that the Cooler Master V550S is able to handle such a card ?

4. I will receive a new card next week, that the shop will have pre-opened and stress tested for a full day before sending it to me. What if the Cooler Master V550S actually destroyed both previous cards, and I actually destroy the next one they send me ? Or am I being silly ?

Ideal scenario would be to start off by using a more powerful PSU once I receive the card, then stress test the card. If it works, then switch over to the V550S. But I mean, I can't send back a third card if I damage it, especially if they tested it for me and made sure it worked beforehand.

Should I simply run the next one they send me (pre-tested to work) with my V550S, and if issues still occur, then simply change the motherboard altogether ?

What do you guys think of all this ?

 

[mfenn]

1. How come when using the Antec 1000W, the graphic card stayed between 65 and 75°c, as opposed to when using the Cooler Master V550S it's at 80°c in full load ?

The Antec 1000W isn't having to work as hard, so is putting out less heat than the VS550. Actually, this is a good hint, perhaps your issue is just that the card is overheating due to insufficient ventilation. Try running with the side of the case off and a desk fan pointed at the card.

2. Both shops I've been to, as well as the computer technician from the company that sold me the card ; they ALL tell me it's my 550W PSU that isn't strong enough. They say that Sapphire recommends a 750W and hence they might now accept taking back my card as they say I may have damaged it with such a low PSU.

They are just trying to get out of helping you and/or sell you more stuff.

Did you measure the voltages like I mentioned in my last post? If the voltages stay within spec or go low, you won't be damaging the card. It's only a problem if the voltages go high for some inexplicable reason.

 

[kokorochin]

Well, I don't have the tools to do that, nor do I know how to.

If I did, I guess I could find tutorials on the web.

But, everything I've done since I got both cards was with the case opened, and the cards didn't overheat, there's 5 leds on the side of the card in order to indicate its temperature, and it's always been 2-3 greens leds max.

I really don't think the issue is from the PSU, because I did try 3 different ones. A stretch could be to assume that the V550S is faulty and damaged both cards as soon as I turned the computer on. I really doubt it though...

I think that either I received 2 faulty cards (super unlikely), or the motherboard is just faulty.

 

[Cerb]

🙁 Mfenn's covered all the main bases.

Are you certain that you did get a 2nd new card? Your Cooler Master PSU may be less than ideal for it, but it should handle it fine, and any 1000W PSU will hardly break a sweat.

I think that either I received 2 faulty cards (super unlikely), or the motherboard is just faulty.

Or, if you returned one card then got another of the same model from the same vendor, that you got your first card re-packaged.

 

[mfenn]

Well, I don't have the tools to do that, nor do I know how to.

A digital multimeter is really cheap if you get a basic one, and something that everybody should have in their toolbox.

 

[kokorochin]

Hi again,

I'm kinda disgusted today by a lot of other stuff but this GPU situation is starting to get me down... I was supposed to receive a card today, but the man in charge have been avoiding my calls and more or less making faulse promises. Anyway, now they tell me 2 weeks waiting time for another R9 280X Toxic... (By the way, I got confirmation that the second card I returned was faulty as well)

Talked with the shop today, they said my situation is so abnormal that they brought it up during a staff meeting. Basically they said their experts decided that I mishandled both cards, that my 550W PSU must have damaged both cards from the get go, so they're treating me like an idiot. They say that this situation happens all the time, that people always say that their less than specified PSU can handle the cards when in fact they can't. I talked with him about watts and consumption, and he said my PSU didn't have enough something. i'm starting to doubt whether or not my PSU is good enough. The fact that the card is 80°C with the 550W and only 75°C with the 1000W shows that something is off, no ?

After 30min of back and forth, and both parties getting pissed off... they were trying to make me buy an R9 290X (way more expensive, I can't afford it).

Now they're offering me the R9 280X Vapor-X. I guess I'll take that one, but I'm really really sad, because I really really really REALLY wanted the R9 280X Toxic, and I even waited so long to get it, let alone find it, and read so many reviews about it...watched so many videos about it, had so much hype about it... Waited more than 4 years to buy my computer and this is what I get...

🙁

 

[mfenn]

I wouldn't get too attached to a fancy overclocked GPU. You can drag a couple sliders to the right on the Vapor-X to get it to the same place.

And no, I don't think that you could have damaged the card by not providing it enough power. The the likely effect of not enough power is that the machine wouldn't power or on that it would power off suddenly while in use (not hang). Artifacting means that the card is defective, end of story.

They're just trying to get out of doing an RMA. The fact that they had a meeting about it and tried to upsell you after the fact indicates to me that something fishy is going on at their end.

 

[kokorochin]

Well,

Fuck that company.

Tired of this situation so considering the many issues with the R9 280X series, I went online and looked at review for another card. My choice got narrowed down to R9 280X Vapor-X or GTX 770 DirectCU II OC.

Great reviews on this card, and a store had some, so I bought an ASUS GeForce GTX 770 DirectCU II OC today.

Works fine with everything, 25min Furmark, 79°c, no problem.

Done !


ps : The guy from the shop told me they have 3 out of their 5 R9 280X Toxic card RMA'ed. Also that one of his friends bought 4, and out of those 4, 2 were faulty. But those who worked, worked really well.

 

[mfenn]

Well there you have it! Bad QC on Sapphire/AMD's part. Sorry that it put you through so much worry. 🙁