|
I mentioned the first link already.
FPS differences in the second link are irrelevant since the FPS values are already so high. As those values come down towards 60 youll notice they even out, and they are pretty much identical anyway. Games where you can get crazy framerates in are good for motherboard benchmarking since thats where you will encounter a motherboard "bottleneck." Or really its where bus limits are tested.
As for BIOS design, subjective really.
The most important part of a motherboard is cost, and south bridge (P,H, or Z). There is only one time someone should consider a motherboard that benches higher is someone who is building an enthusiast rig and already has flagship everything. This is because the money is always better spent elsewhere in the build.
EDIT: Also x16 x4 versus x8x8 (PCI-e lanes)
|
On April 13 2012 09:17 Medrea wrote: I mentioned the first link already.
FPS differences in the second link are irrelevant since the FPS values are already so high. As those values come down towards 60 youll notice they even out, and they are pretty much identical anyway. Games where you can get crazy framerates in are good for motherboard benchmarking since thats where you will encounter a motherboard "bottleneck."
For the most part I would agree with you.
But sometimes the differences in FPS are indicative of performance overall. Here's a good example. Check out the full set of benchmarks for audio and video encoding on the Z68 chipset: http://www.tomshardware.com/reviews/z68a-gd80-p8z68-deluxe-z68xp-ud5,3025-13.html
There's a decent difference between the Asus P8Z68 Deluxe and the MSI Z68A-GD80 and ASRock Z68 Extreme7, yet the AS Rock sells for $10 more on New Egg. and the MSI Z68A-GD80 is $30 cheaper. All the same chipset, yet the board in the middle of the price pack has by far better performance on a real world application.
Now I'll admit things are subjective here, I personally would be willing to pay an extra $30 for a board that has better performance, someone else may not. But that's why it's important to look at benchmarks when purchasing a motherboard.
|
Unless games move over to a transcoding algorithm (for textures) none of those benchmarks matter -.-
|
That's just a matter of default Turbo Boost behavior, which is kind of cheating...? Sometimes there's legit differences in I/O performance because of interface chips used, how they're routed to the chipset, and so on, but there's very little other than possible Turbo Boost shenanigans to distinguish different motherboards in terms of performance. No change in the motherboard is really going to significantly change how the CPU, GPU, memory, and even Southbridge are wired to each other. You're looking at more or less the same logical circuit on every motherboard, for the key components.
Often the differences end up being small enough that it's within the realm of measurement error.
Actually with 7 series chipsets, Asus seems to have reworked the way the memory is interfaced with the CPU a bit, so there should be a small difference. (but again...the memory subsystem, which is really not such a distinguishing factor) http://www.anandtech.com/show/5728/intel-z77-panther-point-chipset-and-motherboard-preview-asrock-asus-gigabyte-msi-ecs-and-biostar/13
|
As far as I'm concerned, motherboard selection goes like this in priority. 1: CPU socket 2: PCIe layout 3: Power phases 4: BIOS quality 5: Warranty and support 6: Chipset cooling 7: Other random feature shit 8: Negligible performance differences that don't mean shit compared to the performance that can be picked up from everything higher on the list.
The real differences from board to board tend to stem from all that other shit, with the exception of warranty. It's design and build quality, good BIOS, and luck of the draw on components.
|
You forgot to mention southbridge. Or is that folded into CPU socket or something.
|
On April 13 2012 12:32 Medrea wrote: You forgot to mention southbridge. Or is that folded into CPU socket or something.
Well, it kinda crosses PCIe layout and other random features. At least in the ways I factor it. It's also not a precise prioritization, just the rough layout.
The general idea is that roughly everything else is more important than "motherboard benchmarks".
Things that aren't just as likely to be related to luck of the draw on components on the individual sample, but cold, hard features.
|
On April 13 2012 12:20 JingleHell wrote: 8: Negligible performance differences that don't mean shit compared to the performance that can be picked up from everything higher on the list.
I guess I'm just far more anal about performance than everyone else
|
On April 13 2012 13:01 TheToast wrote:Show nested quote +On April 13 2012 12:20 JingleHell wrote: 8: Negligible performance differences that don't mean shit compared to the performance that can be picked up from everything higher on the list. I guess I'm just far more anal about performance than everyone else
Or looking for it in a different way? I expect to put some effort into the performance. Time spent fucking around in the BIOS, benching, stressing, tweaking minute shit. Clean power delivery, a good BIOS, and solid chipset cooling will make more difference to the end result than anything else. I'm not willing to sacrifice quality of life and warranty for "gains" that are well inside the margin of error and individual sample differences. Also, performance on a bench in a climate controlled lab is one thing, performance I can get inside an air cooled case is potentially different. There's too many factors that play into it to just go by benchmarks where the motherboard is concerned.
|
Being anal about performance is one thing. Being anal about performance per dollar is an entirely different thing altogether.
|
On April 13 2012 13:11 Medrea wrote: Being anal about performance is one thing. Being anal about performance per dollar is an entirely different thing altogether.
Nah, I generally pay way more attention to the performance side than the dollar side and still think mobo benches are generally a bit silly compared to other shit.
|
I figured out recently that some of my friends play SC2 and we wanted to LAN together this summer. I was looking for laptops for college, and was recommended these two (which I have no idea what the difference is between the two, they seem nearly similar, just a $100 price tag difference between them). http://www.newegg.com/Product/Product.aspx?Item=N82E16834246221 http://www.newegg.com/Product/Product.aspx?Item=N82E16834246324
Are these still ideal laptops to purchase? I'm planning on running SC2 on the lowest settings possible. I intend to do 4v4 matches, and would not want to lag during big fights. Would either of these laptops be worth the money or even handle what I am looking for? Side note: does it make a difference if I'm going to use an external monitor while using this laptop for gaming? I don't know much about resolution, but if I use a larger monitor..would my FPS go down?
|
The more expensive one has a quad-core rather than dual-core processor. That's pretty much the only difference aside from the hard drive size (500GB on the more expensive one, compared to 750GB). For what you're looking at, a quad core is probably not worth it.
The price is not particularly good on either.
If you use an external monitor with a resolution higher than 1366 x 768 (the laptop's resolution), which is pretty much any monitor you'd be using, and run a game on the monitor's native resolution, then your fps would go down in most games. If there's more pixels for the graphics processor to handle, then it's not going to be able to render frames as quickly, hence the lower fps.
Every single computer in the world can lag a bit on 4v4, but something newer is most likely going to run that better than whatever you're using now.
For college I'd consider prizing durability and maybe weight and battery life over gaming performance, but it's your choice. Then again, it's hard to get something of really high quality in the $800 range, for a laptop.
|
Is there a reason that the i7-2600k is so cheap ($200) right now at microcenter? For comparison, the i5-2500k is still at $180 and the i7-2700k is at $300. Would this be a worthwhile $20 upgrade over an i5-2500k in terms of longevity? Is hyperthreading worth $20 to the average user?
|
If you can get an i7-2600k for $20 on the i5-2500k I would say that yes it is worth it. It offers no gaming performance, but its still worth it considering thats only the cost of a pizza dinner.
|
Clearing stock. 2600k and P8Z68-V for under $300, pretty amazing - well worth it.
|
|
No, I thought it was the P8Z68-V since that's what the OP on [H] mentioned but I guess he's an idiot. But LX variant for $70 is okay and would be worth it for most people.
|
Does anyone know how to change versions of Ubuntu. I have (I think) the version that is optimized for AMD cpu's on my old HDD, and I'm using an Intel motherboard and processor, so I'd like to change this if it doesn't mean losing all the work I've already done within Ubuntu. Thanks!
Also if anyone knows of a way to test my performance in Ubuntu, I'd be grateful to know.
|
Do you know if the LX variant is missing any overclocking tools? I have an ASUS p8p67-M that won't let me manually adjust vcore (though I guess offset works too).
This mobo also applies to microcenter's $50 off combo, is it worth the extra $30 over the ASUS LX? http://www.microcenter.com/single_product_results.phtml?product_id=0364085 I know skyR prefers ASUS over Gigabyte.
|
|
|
|