EPTAnd if you've never worked inside a prebuilt before. Getting the power supply out is going to be a daunting task.
Computer Build Resource Thread - Page 830
| Forum Index > Tech Support |
When using this resource, please read FragKrag's opening post. The Tech Support forum regulars have helped create countless of desktop systems without any compensation. The least you can do is provide all of the information required for them to help you properly. | ||
|
skyR
Canada13817 Posts
And if you've never worked inside a prebuilt before. Getting the power supply out is going to be a daunting task. | ||
|
scarymeerkat
Canada107 Posts
| ||
|
Pawsom
United States928 Posts
What is your budget? $1200 US. This is flexible, if his goals are possible without spending this much, he would surely prefer to save money or possibly get an SSD. What is your resolution? Unsure as the actual resolution, but I know he wants a 15-16 inch screen. He has used a 17 inch and finds it too large. What are you using it for? One thing my friend mentioned is important to him is a laptop that doesn't get too hot if its on his legs. Use will be gaming(WoW, SC2, COD, etc) and typical everyday use. What is your upgrade cycle? It needs to last for 4 years. When do you plan on building it? Plan on buying it asap. Do you plan on overclocking? No Do you need an Operating System? I'd assume one comes with a laptop? If not then I'm not sure, but don't let it factor much into the budget. Do you plan to add a second GPU for SLI or Crossfire? No Where are you buying your parts from? I'd like to order the computer online or from a retail store near Detroit, MI. Thanks guys  If theres a better place to ask about laptops please let me know | ||
|
skyR
Canada13817 Posts
On December 26 2011 12:17 scarymeerkat wrote: Right. That makes sense. It WOULD however be doable? Almost definitely not worth it though. And I'm assuming (from the minimal research I've done) that those graphics cards would be absolutely awful options, even if he's not terribly concerned with settings? Yes it's doable and yes both the GT420 and GT520 are terrible (except for if you want to play on low settings at a low resolution). | ||
|
Myrmidon
United States9452 Posts
On December 26 2011 11:58 Wabbit wrote: Ï know you know way more than I do, so in the interest of learning, may I ask for a little of your time to explain: "PSHII is individual voltage regulation and has a few less quirks maybe" (emphasis mine) I think you're hinting at the possibility of crossloading here with the "maybe", not sure when that would happen in a modern system though. Also from the reviews I've seen of the PSHII, the voltage regulation is actually worse. TweakTown's review of the GS800 (assuming it's not flawed and that it transfers to the 600) shows it actually goes out of spec on 5V, dropping too low. Ripple is also higher on the GS/PSHII than the S12II. The only issue I see as worrisome with the Antec is the fan. + Show Spoiler [long response] + Well the "maybe" was talking about the sometimes funny transient response of S12II. Crossload is of course not practical. It's poor here: http://www.techpowerup.com/reviews/Jou_Jye/BW-B520JL/6.html http://www.techpowerup.com/reviews/Antec/HCG-620/6.html Great here, somehow: http://www.techpowerup.com/reviews/Antec/HCG-520/6.html By theory (not looking at test results), group regulation should be at a significant disadvantage compared to individual regulation for real-world scenarios where +12V spikes very quickly while the other rails don't change too much. TechPowerUp's transient load test probably loads +3.3V and +5V more than is realistic, but that is a good way to coax out problems. It's arguably a +3.3V/5V heavy crossload. But an almost pure +12V spike would be similarly tough for a group regulated unit to handle. Then again, the GS800 didn't do so well either. http://www.techpowerup.com/reviews/Corsair/GS800/6.html But that's mostly because the minor rails don't regulate that well in general. With a +12V-only transient, I wouldn't expect such bad results. Comparing one platform at say 520-620W compared to another at 800W is not fair, as the lower-wattage variants of the same platform tend to have better voltage regulation and ripple. It's generally fairer to look at the 800W unit's performance at the lower load like 600W than to assume the 800W unit's full load results will be like the 600W unit's full load performance. Anyway, I wouldn't take minor rail voltage regulation too seriously if they test the rails at very high settings, as computers generally don't draw that much from those. Testing units to their full 3.3V and 5V capacity is well and good, but you're not going to see those numbers in practice. See the wattage from this system, for example: http://www.xbitlabs.com/articles/cases/display/system-wattage_8.html#sect0 Granted, each hard drive can draw about 1-3W from +5V (0.2-0.6A), and so on, so if you've got a stack of those, you could get a little higher than the above system with one hard drive. Realistically, the fan may be the main concern, since users aren't likely to notice anything else. On December 26 2011 12:12 scarymeerkat wrote: http://www.futureshop.ca/EN-CA/product/id/10164624.aspx http://www.futureshop.ca/EN-CA/product/id/10180942.aspx Seems inefficient that he's asking through me but whatever. My friend is looking at these two computers. Obviously the graphics cards are pretty terrible. He's asking if he could just replace the graphics cards for these, because apart from that he thinks they look pretty tempting. Is that possible? Any reasons why (if that was possible) that would be a bad idea? I still think the build looks better, but what do I know? It's not tempting because the prices aren't so special. Moreover, for a reasonable graphics card upgrade, he'd have to replace the power supply, which would require almost as much work (and stress if there is any) as putting together the whole computer, anyway. These OEM boxes are not really designed for airflow paths that will cool a higher-end graphics card. It would still be fine, but it's not optimal. | ||
|
scarymeerkat
Canada107 Posts
And thanks for helping me with my computer a while back. Still greatly appreciated. | ||
|
PigAntlers
Canada32 Posts
I'm looking into getting a new mobo and cpu while i wait for next gen gpu's to come out I made a post of my specs back on page 801. I got win 7 64 bit but before I upgrade from 32 bit Vista I want to get my new mobo and cpu because I read their can be errors if you don't do it that way. I have been looking into i5 2500k's (I will like to OC) and i see a couple packages on NCIX atm with 2 different mobo's and I know NOTHING about mobo's like what makes one better than another. I have 2 displays a tv and my 23" LG monitor and I will be upgrading my Ram too which I dont know of a big difference between different brands, and I will most likely be upgrading to a single powerful gpu but I would like to keep crossfire in mind or sli just in case. My comp will be a high end powerhouse when I'm done. So http://ncix.com/products/index.php?sku=64195&vpn=i5 2500K&P67X-UD3-B3&manufacture=Bundle Deals&promoid=1301 or http://ncix.com/products/index.php?sku=62550&vpn=2500K & P67A-C43&manufacture=Bundle Deals&promoid=1301 Or if theirs a cheaper way to go about this like not getting these packages | ||
|
skyR
Canada13817 Posts
Upgrading from Vista to 7 does not require new hardware... whoever told you this is an idiot. | ||
|
Wabbit
United States1028 Posts
On December 26 2011 12:31 Myrmidon wrote: + Show Spoiler [long response] + Well the "maybe" was talking about the sometimes funny transient response of S12II. Crossload is of course not practical. It's poor here: http://www.techpowerup.com/reviews/Jou_Jye/BW-B520JL/6.html http://www.techpowerup.com/reviews/Antec/HCG-620/6.html Great here, somehow: http://www.techpowerup.com/reviews/Antec/HCG-520/6.html By theory (not looking at test results), group regulation should be at a significant disadvantage compared to individual regulation for real-world scenarios where +12V spikes very quickly while the other rails don't change too much. TechPowerUp's transient load test probably loads +3.3V and +5V more than is realistic, but that is a good way to coax out problems. It's arguably a +3.3V/5V heavy crossload. But an almost pure +12V spike would be similarly tough for a group regulated unit to handle. Then again, the GS800 didn't do so well either. http://www.techpowerup.com/reviews/Corsair/GS800/6.html But that's mostly because the minor rails don't regulate that well in general. With a +12V-only transient, I wouldn't expect such bad results. Comparing one platform at say 520-620W compared to another at 800W is not fair, as the lower-wattage variants of the same platform tend to have better voltage regulation and ripple. It's generally fairer to look at the 800W unit's performance at the lower load like 600W than to assume the 800W unit's full load results will be like the 600W unit's full load performance. Anyway, I wouldn't take minor rail voltage regulation too seriously if they test the rails at very high settings, as computers generally don't draw that much from those. Testing units to their full 3.3V and 5V capacity is well and good, but you're not going to see those numbers in practice. See the wattage from this system, for example: http://www.xbitlabs.com/articles/cases/display/system-wattage_8.html#sect0 Granted, each hard drive can draw about 1-3W from +5V (0.2-0.6A), and so on, so if you've got a stack of those, you could get a little higher than the above system with one hard drive. Realistically, the fan may be the main concern, since users aren't likely to notice anything else. Thanks for taking the time. I actually hadn't seen the transient tests [for those units] before. It's interesting to see the same platform fail in one test and pass in another. I'm actually kind of disappointed that more PSU reviewers (who otherwise seem to have the proper equipment and methodology) don't do transient load testing. Be assured that I took the time to actually read it all, you didn't waste time typing all that up | ||
|
PigAntlers
Canada32 Posts
| ||
|
skyR
Canada13817 Posts
| ||
|
Myrmidon
United States9452 Posts
On December 26 2011 12:47 Wabbit wrote: Thanks for taking the time. I actually hadn't seen the transient tests [for those units] before. It's interesting to see the same platform fail in one test and pass in another. I'm actually kind of disappointed that more PSU reviewers (who otherwise seem to have the proper equipment and methodology) don't do transient load testing. Be assured that I took the time to actually read it all, you didn't waste time typing all that up That's the weird thing to me, how results can be that different. Is it sample variance? Different batch of components used? Capacitor or inductor values maybe being way off of nominal values? Usually when high tolerance components are used (big electrolytic caps are often rated at like +- 20% of nominal), the exact value doesn't really matter, either from the nature of the function they perform, or from the designers being clever and designing around that potential variance. I really think a key result that you can see in the xbitlabs link, is low little +3.3V and +5V draw changes for a system. It's almost all +12V that changes at different loads. Granted, I think that +3.3V and +5V may come down more at idle as opposed to "Windows startup." | ||
|
Wabbit
United States1028 Posts
On December 26 2011 12:59 Myrmidon wrote: That's the weird thing to me, how results can be that different. Is it sample variance? Different batch of components used? Capacitor or inductor values maybe being way off of nominal values? Usually when high tolerance components are used (big electrolytic caps are often rated at like +- 20% of nominal), the exact value doesn't really matter, either from the nature of the function they perform, or from the designers being clever and designing around that potential variance. I really think a key result that you can see in the xbitlabs link, is low little +3.3V and +5V draw changes for a system. It's almost all +12V that changes at different loads. Granted, I think that +3.3V and +5V may come down more at idle as opposed to "Windows startup." I was going to speculate about the same thing but in a far less articulate way :D I think the biggest conclusion to draw is that for people with high-end/OC'd CPU /GPU and especially multi-GPU, it's important to check transient loads for 12V and see how the PSU they're considering performs when going from like 10-20% load to a quick additional 20+A on the 12V (though atm I'm hving a hard time thinking of a scenario where you'd get that load for just a little bit :D Maybe zipping a large file on an OC'd bulldozer? EDIT: Though off to full on 12V and VSB to full on 12V are realistic, at least JohnyGuru does ém) | ||
|
ssartor
United States129 Posts
| ||
|
skyR
Canada13817 Posts
| ||
|
Myrmidon
United States9452 Posts
| ||
|
-bambi-
Canada11 Posts
-Twin Frozr 6950 (http://ncix.com/products/index.php?sku=64182&vpn=R6950%20Twin%20Frozr%20III%201G%2FOC&manufacture=MSI%2FMicroStar&promoid=1301) -Gigabyte non-ti GTX 560 (http://ncix.com/products/index.php?sku=61236&vpn=GV%2DN56GOC%2D1GI&manufacture=Gigabyte&promoid=1301) -Twin Frozr 560ti 448 core edition (http://ncix.com/products/index.php?sku=65880&vpn=N560GTX%2DTi%20448%20Twin%20Frozr%20III%20PE%2FOC&manufacture=MSI%2FMicroStar&promoid=1301) There is quite a price difference between the 3, so I was wondering which would be the best at this point. Cheers | ||
|
Ata
Canada356 Posts
On December 26 2011 11:23 Myrmidon wrote: Yes it fits, but almost always I'd recommend getting a wireless adapter that's not stuck to the back of the case where reception may be blocked by the chassis or just poor anyway. Reception is usually worse lower to the ground, anyway. Adapters with external antennas are therefore preferred. For USB models you can use a M/F USB extension cord if one is not supplied, and raise it up somewhere. So this or maybe something with two antennas if the 802.11n router supports two spatial streams and the extra data rate is actually needed (for most people no, since they're just transferring to outside of the home network, where the link to the ISP is the bottleneck by far). http://ncix.com/products/?sku=44688&promoid=1301 Thank you very much, also a big thank you to SkyR. Have a great holidays. | ||
|
skyR
Canada13817 Posts
On December 26 2011 14:14 -bambi- wrote:+ Show Spoiler + So now that the deals on GTX 560ti cards are all sold out at NCIX, I'm trying to decide between: -Twin Frozr 6950 (http://ncix.com/products/index.php?sku=64182&vpn=R6950%20Twin%20Frozr%20III%201G%2FOC&manufacture=MSI%2FMicroStar&promoid=1301) -Gigabyte non-ti GTX 560 (http://ncix.com/products/index.php?sku=61236&vpn=GV%2DN56GOC%2D1GI&manufacture=Gigabyte&promoid=1301) -Twin Frozr 560ti 448 core edition (http://ncix.com/products/index.php?sku=65880&vpn=N560GTX%2DTi%20448%20Twin%20Frozr%20III%20PE%2FOC&manufacture=MSI%2FMicroStar&promoid=1301) There is quite a price difference between the 3, so I was wondering which would be the best at this point. Cheers 6950 Twin Frozr III. Gigabyte Windforce is worse than Twin Frozr III and has no RMA center in Canada. The GTX 560 Ti 448 Twin Frozr III just isn't hot. | ||
|
Divinek
Canada4045 Posts
And is the 500$ budget one from this thread http://www.teamliquid.net/forum/viewmessage.php?topic_id=271861 most likely the best deal for that price ill be able to get from canada? And will it most likely be even cheaper on boxing day? | ||
| ||
