Page 2 of 3 FirstFirst 123 LastLast
Results 21 to 40 of 44
  1. #21
    I don't rate Dell's at all really. They have some really good selling points (good processor, great storage), but I think the graphics card is a bit naff, and in terms of upgrading you have no where to go really. They are very tightly compacted together inside, so you can't really put anything exciting in there. Also, they cut it really fine with the PSU (460W last time I looked), just rules out putting a really meaty graphics card in there.

    The Chillblast looks quite good, have you had a look at somewhere like PC Specialist or Scan?

  2. #22
    Secondary Hivemind Nexus Sakkura's Avatar
    Join Date
    Jul 2012
    Location
    Denmark
    Posts
    1,410
    Quote Originally Posted by BiilyTheKid View Post
    The Chillblast looks quite good
    No, it really really doesn't.

  3. #23
    Activated Node
    Join Date
    Aug 2011
    Posts
    88
    My goodness, all this about SSDs and HDDs is rather silly considering we're talking about a desktop.

    If we were on about a laptop with only one drive bay, I could see the point, but on a desktop I can't. A new build should be fine with an SSD of about 256GB (heck, even 128GB). If this is a first desktop, then you just have to make do with the limited space until you can afford a HDD to compliment it (should be enough for a fair few games and other stuff). That shouldn't be much of a problem as 1TB 3.5" HDD are very cheap, ridiculously cheap in fact. If this is a replacement desktop, then you can use your old HDD and then if you really want buy another HDD later. Either way, you get the benefits of an SSD sooner rather than later.

  4. #24
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    It's rather silly considering the unnecessary assumption of LowKey's preferences; LowKey can decide what storage is necessary and purchase accordingly given the suggested options.

    @Sakkura: I dunno, for gaming and general use 1333 vs 1600 RAM is irrelevant. The 3570 thing is probably just a mislabeling, which is perhaps unprofessional but wouldn't affect performance. I sent them an email about the CPU in this unit, which I think is the one LowKey was looking at (although it had Windows 7 instead of 8):

    http://www.chillblast.com/Chillblast-Fusion-Blaze.html

    Aside from the questionable PSU, that doesn't look unreasonable. I am quite worried about EZCool though; many a report of overrating. It'd have to be super-overrated to not handle that PC even with the overclock, but overrating is also usually associated with crappy components and poor QA...

  5. #25
    Secondary Hivemind Nexus Sakkura's Avatar
    Join Date
    Jul 2012
    Location
    Denmark
    Posts
    1,410
    There is some scaling from DDR3-1333 to 1600. They also cost nearly the same because 1600 is standard today. So there's no reason to go 1333.


  6. #26
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    FPS is a pretty poor measure of the visual consistency of GPU output, as explicated by Scott Wasson of the Tech Report in the previous three links and by Ryan Shrout of PC Perspective in this one. In any case, I don't think 1.5 frames of difference is "relevant".

    Marginally-faster RAM is nice, but it's not necessary for gaming and general use. I'm unfamiliar with UK shops so I just judge what's presented, and the listed ChillBlast unit (if the power supply doesn't fry things) would be fine.

    tbh I'm more offended by the brand name (Chillblast, really?) than I am worried about the components (PSU aside). That's almost as damned-silly as MSI's "Twin Frozr" marketing, but it's the name of the whole company.

    Of the two, I'd still lean towards the Dell because, as I said, they usually have overbuilt (if older) PSUs.

  7. #27
    Secondary Hivemind Nexus Sakkura's Avatar
    Join Date
    Jul 2012
    Location
    Denmark
    Posts
    1,410
    Quote Originally Posted by Jambe View Post
    FPS is a pretty poor measure of the visual consistency of GPU output, as explicated by Scott Wasson of the Tech Report in the previous three links and by Ryan Shrout of PC Perspective in this one. In any case, I don't think 1.5 frames of difference is "relevant".
    It's more than a 5% performance difference, and that's not likely to be distributed evenly, ie. it helps prevent FPS spikes or stuttering and thus most likely represents a larger performance difference in practice.

    And it's basically free.

  8. #28
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    Quote Originally Posted by Sakkura View Post
    It's more than a 5% performance difference
    Yes, and if the benchmarks were ten frames slower it'd be 7%, and twenty frames slower would be 15%, and so on. A frame and half of difference is irrelevant.

    Quote Originally Posted by Sakkura View Post
    it helps prevent FPS spikes or stuttering and thus most likely represents a larger performance difference in practice.
    I sincerely doubt that is the case, but I know of no studies to that effect. Do you?

  9. #29
    Secondary Hivemind Nexus LowKey's Avatar
    Join Date
    Jun 2011
    Location
    Edinburgh
    Posts
    1,212
    Thanks for discussion guys very interesting, I have thought about storage actually as I am currently using a fairly limited 500GB, think I will stick with the conventional storage for now as I cant quite stretch to both, think I will stick to the first chillblast one and swap in a better PSU and ram which shouldn't cost too much more.

    As for the name, very much agree, looking at one at DinoPC and leaning there just for the name (although its not much better), any recommended retailers would of course be appreciated, thanks again!

  10. #30
    Secondary Hivemind Nexus Sakkura's Avatar
    Join Date
    Jul 2012
    Location
    Denmark
    Posts
    1,410
    Quote Originally Posted by Jambe View Post
    Yes, and if the benchmarks were ten frames slower it'd be 7%, and twenty frames slower would be 15%, and so on. A frame and half of difference is irrelevant.
    No, a 5% difference is not irrelevant, and it's not a fluke since the pattern is very clear.

  11. #31
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    @LowKey: heh, I was just bemoaning silly marketing. MSI's Twin Frozr graphics cards are actually pretty good.

    There may well be better pre-built computers available, but I know nothing about UK PC retailers...

    A good 400W PSU would be plenty for a 3570/GTX 660 with the overclocks; the Corsair CXM 430W for £40 is good. If you think you'll ever upgrade to a beefier GPU, you'd probably need two PCI-E power connectors, so you might go ahead and get a PSU with two PCI-E power connectors (e.g. the CXM 500W (£50)). There are non-modular versions of those that are a bit cheaper. There are probably other good Bronze-rated PSUs in that price range, but most of the PSUs I see at Amazon are unfamiliar to me.

    @Sakkura: if one cannot discern 28.5 FPS from 30 FPS, then the difference does not matter. If you actually know of evidence which would indicate that a human being can discern such a difference, or that 267Mhz of RAM speed would improve gaming or general use performance in any way that is meaningful to an end user, I'd gladly read it. Otherwise we're just volleying an inconsequential triviality.

  12. #32
    Secondary Hivemind Nexus Sakkura's Avatar
    Join Date
    Jul 2012
    Location
    Denmark
    Posts
    1,410
    It's not 267 MHz, it's 133 MHz.

    Anyway, if you apply the same logic about 5% across all components, you'd end up with much more than a 5% difference, and then the difference would be immediately obvious even if 5% isn't (though it probably is).

  13. #33
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    Well, 133.3~ if you want to be technical, but yes, DDR.

    I don't know what you're trying to say with that last bit. RAM effects every component in the system and human beings can "probably" tell 28 FPS from 30?

    Again, I've no reason to feel this difference in RAM speed is meaningful if one's just gaming. I've never actually read a controlled analysis of this particular issue, though; I could be wrong. I find what you're positing extremely unlikely.
    Last edited by Jambe; 06-05-2013 at 08:59 PM.

  14. #34
    Secondary Hivemind Nexus LowKey's Avatar
    Join Date
    Jun 2011
    Location
    Edinburgh
    Posts
    1,212
    Righto a few revisions and a lean towards DinoPC has left me with -
    CPU: 2085 - Intel Core i5 3570K
    CPU Overclocking: oc43 - 4.3Ghz - Moderate Overclock
    CPU Cooler: 3019 - Xigmatek Dark Knight Nighthawk SD1283
    Operating System: 11014 - Microsoft® Windows 7 Home Premium 64-bit
    Motherboard: 4164 - Asus P8Z77-V LX2
    RAM: 5041 - 8GB Corsair 1600mhz Vengeance (2x4GB)
    Hard Drive: 6029 - 2TB S-ATAIII 6.0Gb/s
    Optical Drive: 7003 - 22x DVD±RW DL S-ATA
    Graphics card: 8087 - NVIDIA GeForce GTX 660 2GB
    Sound card: 10001 - Onboard 7.1 Audio
    Internet: 13006 - Wireless 802.11N 300Mbps MIMO PCI card
    Keyboard and Mouse: 14001 - Keyboard, mouse and speakers
    Monitors: 15043 - LG 21.5" 22EN33S-B LED
    Case: 17024 - Xigmatek Asgard
    PSU: 18046 - 550W Corsair VS
    Warranty: sc-1 - 3 Year SureCare Warranty

    Going to get the monitor on ebuyer which will save me enough to cover the delivery, totalling £850, which is as much as my poor wallet can take unless there are any game breakers.

    Edit: should note that the overclocking is included without cost.

  15. #35
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    Looks alright. Corsair's VS series PSUs appear to be old and rather inefficient designs (being the bottom tier of the company's offerings), but they're put together by (post-QA-tuneup) CWT and should at least be stable (which is more than can be said for the other computer's EZCool unit, which is a total crapshoot afaict).

  16. #36
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    fwiw, Thom from sales@chillblast.com says:

    The Processor would be the K variant. Apologies for the typo on the website.

    Kind Regards

  17. #37
    Secondary Hivemind Nexus Sakkura's Avatar
    Join Date
    Jul 2012
    Location
    Denmark
    Posts
    1,410
    Quote Originally Posted by Jambe View Post
    Well, 133.3~ if you want to be technical, but yes, DDR.

    I don't know what you're trying to say with that last bit. RAM effects every component in the system and human beings can "probably" tell 28 FPS from 30?

    Again, I've no reason to feel this difference in RAM speed is meaningful if one's just gaming. I've never actually read a controlled analysis of this particular issue, though; I could be wrong. I find what you're positing extremely unlikely.
    Humans are surprisingly good at distinguishing different frame rates. Even more so when stuttering is involved, which it may very well be where memory is concerned.

    RAM speed is meaningful in games, it's just a matter of where you cut off. A modern computer with EDO RAM would perform horribly. The gap between DDR3-1333 and 1600 is not huge, but neither is the gap between 1333 and 1066, and so on. So you could use the same argument recursively until you reach EDO RAM (or worse). The point is you need to draw the line somewhere, and since DDR3-1600 is a bit better than DDR3-1333 while costing practically the same, it's the most logical place.

    Quote Originally Posted by Jambe View Post
    Looks alright. Corsair's VS series PSUs appear to be old and rather inefficient designs (being the bottom tier of the company's offerings), but they're put together by (post-QA-tuneup) CWT and should at least be stable (which is more than can be said for the other computer's EZCool unit, which is a total crapshoot afaict).
    Yeah, I also think it looks okay. Corsair VS isn't actually old at all, it's just their bottom-rung series intended for poor countries (India and such). Corsair CX would normally be their cheapest set of PSUs in our part of the world.
    The VS units are 220-240V only, so they'll never show up in the US, but I guess they figured Scottish people are Scrooges and brought them in.

  18. #38
    Network Hub Jambe's Avatar
    Join Date
    Jun 2011
    Location
    Indiana
    Posts
    310
    I agree that you might as well get 1600, but it's not a major purchase point...

    Anecdotal evidence from years poking at and in computers leads me to believe it's unnoticeable, but if you know of perceivable knock-on effects emerging from this tiny difference in RAM, I'm sure the likes of Scott Wasson et al. (who make a living from pointing out such things) would be interested in a note. What little study I've seen of RAM speed's correlation to perceived visual smoothness in games (from TR, PCPer, and Anand) indicate that the difference between 1333 and 1600 is unnoticeable (tests were conducted with a mainstream GPU and 60hz 1080p monitors iirc).

    Corsair's VS series are rebadged Channel Well GPA platforms; they've been around since early 2011. They're the cheapest platform I'd personally feel comfortable using, and I wouldn't recommend such a unit if one was buying a PSU by itself, but it's not bad by any means. Corsair insists on more stringent QA on their CWT-built units, which is a bonus; the VS series probably has caps from the more reputable Taiwanese and/or Chinese suppliers (as opposed to just "whatever they can get").

  19. #39
    Activated Node KilgoreTrout_XL's Avatar
    Join Date
    Jan 2012
    Location
    Cambridge, Mass.
    Posts
    45
    Good luck with the build- that ought to be a great rig.

    My only suggestion would be to swap out the Asgard for... anything, really. There's a reason it's so cheap. I went shopping for a case and rebuilt my pc just this past weekend, and the asgard was on display at the local store. Everything about it seemed flimsy, from the metal enclosure to the drive bays. The tool-free design was a horror show. The sales guy mentioned that people had reported having a hell of a time with installing and securing their motherboards (and it was noisy as hell).

  20. #40
    Secondary Hivemind Nexus LowKey's Avatar
    Join Date
    Jun 2011
    Location
    Edinburgh
    Posts
    1,212
    Ohh interesting, I just went with it as it was the plainest :), I'll look at something a bit smarter.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •