Its a 30nm process node vs I think 40nm for all other RAM. That means they are physically smaller, run cooler and faster with less voltage
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
"I need a New PC!" 2012 Thread. Ivy, SSDs, and reading the OP. [Part 2]
- Thread starter Hazaro
- Start date
- Status
Its a 30nm process node vs I think 40nm for all other RAM. That means they are physically smaller, run cooler and faster with less voltage
Horse Detective
Why the long case?
Putting together a rig now after an offer on my current computer.
Not going to order this without a monitor. I wonder if I would notice a huge difference between my current ips screen, and led options such as those offered from Samsung. They have a 23" on new egg from about 169...Does Gaf have a preferred monitor choice
Not going to order this without a monitor. I wonder if I would notice a huge difference between my current ips screen, and led options such as those offered from Samsung. They have a 23" on new egg from about 169...Does Gaf have a preferred monitor choice
Horse Detective
Why the long case?
Ooh, do not want that. I guess crossover it is then. Dell IPS monitors are considerably more expensive, so I guess that is a no go.
This world of custom PCs is scary, and complicated.
edit: Ohhhh that Achieva Shimian is beautiful.
Horse Detective
Why the long case?
Thank you for the recommendation on the Crossover 27q. I am trying very hard to contain drool while watching videos of it.
Can anyone help with this?
If you want to get a good look at the panel, just look at your iMac. It is literally the same one.
AMD's working on the drivers to fix the microstuttering. Apparently, it's something to do with GCN's memory management.
http://techreport.com/news/24136/dr...ame-latency-troubles-to-be-fixed-with-updates
I was wondering why they hadn't released anymore beta drivers to increase FC3 performance.
http://techreport.com/news/24136/dr...ame-latency-troubles-to-be-fixed-with-updates
I was wondering why they hadn't released anymore beta drivers to increase FC3 performance.
Koroviev
Member
Installed the new hardware today. Took way longer than it should have owing to the stupid proprietary front panel wiring, but luckily I was able to work around it. The case is not ideal, but my next priority is the GPU.
I'm currently working with a GTX 460 (768MB)...it has been a very solid card, especially considering that I only paid $80 for it new. However, I would like to pick up a newer card in the near future, and I'd like to spend $150-$200.
So...when is a good time to buy, and which models should I be looking at? My monitor is 1080p.
I'm currently working with a GTX 460 (768MB)...it has been a very solid card, especially considering that I only paid $80 for it new. However, I would like to pick up a newer card in the near future, and I'd like to spend $150-$200.
So...when is a good time to buy, and which models should I be looking at? My monitor is 1080p.
D
Deleted member 22576
Unconfirmed Member
I'm doing it. I'm getting a 21:9 monitor and its gonna be awesome. More ostentatious and bigger than a 24inch standard panel, less hassle than dual display, and not quite the resource drain of a 2k+ jumbotron. Its the perfect solution.
I've had my computer for about 3 months now and its still just plugged into my tv in a ghetto fashion.
I've had my computer for about 3 months now and its still just plugged into my tv in a ghetto fashion.
Koroviev
Member
Huh, that's interesting. I've just read that I should be looking at 1GB+ vram for my next card. I was fine getting the 768mb model originally because the difference between it and the 1GB model was generally very insubstantial with respect to benchmarks. I also looked at min/average charts and I wasn't seeing any alarming fps drops.
Edit: double-post, sorry
LiquidMetal14
hide your water-based mammals
Performance per dollar has to be accounted for though. Some of the AMD GPU's are a far better proposition.
I just saw a 670 for 400 and that seems like a great deal and not far off from a 680. I'm basing my next system off of that GPU as a matter of fact.
On a side note - I decided to go with the h100i water cooler. Thanks to mkenyon's always handy advice, I decided to go with with the Corsair 500R. I really liked this case and a side bonus is it has a top cooling intake/outtake spot for dual fans or basically one of their h100 water coolers.
The only remaining question regards air flow. If I'm using the radiator with fans on the top, is it best to use the top fans as intake or outtake? And if that does change, should I change any of the other case fans like the one in the back to an in or out? This all seems simple but air flow is serious business /headache
Koroviev
Member
Are people having good experiences with the current generation of amd cards? Solid driver support and all that? I didn't have a great experience with my last amd card, but it has been awhile.
Most are (as you'll probably hear from this forum), but if you really want Nvidia then the 660 is a decent card. Anything cheaper than the 660 like a 650 ti or just 650 is not really worth it, same with Radeon's 7770 and 7750 unless you need the 7750 for its low power draw.
Problems with the 660 is that it is lacking in RAM (not a true 2GB card), somewhat lacking in memory speed and doesn't overclock as well as the 7850. The 78XX overclock really well and the 7850 overclocks the most over stock of the two because it has the lowest base clock of the two.
LiquidMetal14
hide your water-based mammals
Their drivers are not up talk to par with nvidia but when you talk meat and potatoes performance, they are excellent options.
The only thing I would say for nvidia is that they have good driver support with PhysX support and custom AA solutions that AMD hasn't quite matched. Performance is higher but at a price.
That's why I recommend a 670 if you're going to go for an nvidia card unless you can splurge and spend the extra 50-150 on a nice 680.
He was saying 150-200 for a new card. Don't know if he would like to double the budget.
Couchaddict
Member
Back in the OP, it recommends the SS sensai has best mouse. Anyone has any experience using claw grip on it?
Long shot : Any lefty claw grip user?
I currently have a Logi G3 but I need a new mouse at work, this Dell BS is destroying my hand.
Long shot : Any lefty claw grip user?
I currently have a Logi G3 but I need a new mouse at work, this Dell BS is destroying my hand.
Question. We need a card that has HDMI out for a PC at my work. I can take this opportunity to upgrade my current card at home, as it has HDMI out, and use the new card at home. That said, I don't have a crazy budget, and I need to have it today, which is why I'm linking to Best Buy.
So, is this GTX 660 a worthwhile upgrade over my current GTX 465?
Thanks.
So, is this GTX 660 a worthwhile upgrade over my current GTX 465?
Thanks.
Koroviev
Member
I'd be okay with a 5xx card. I'll probably upgrade it in a year or less anyway.
And yeah, I'd rather not spend more than I spent on the i5 processor since I upgrade my GPU yearly. I don't sell my old cards since my younger sister can use them.
That's because Min/Max/Average is a really bad way to measure performance. For example, here's 3 minutes of gameplay in Firefall. First chart is min/max/average, second chart is a plotline of the FPS.
Minimum 83 FPS, pretty good right? Wrong. Here's a plotline of the exact same data that shows, in milliseconds, how long it took to render each frame.
There are a huge number of frames that are taking over 16.7ms (60fps) to render, with some reaching 70ms (14fps). These are the kind of stutters that one would encounter when running out of VRAM.
A much more accurate measurement of 'average' experience is 99th Percentile Frame Latency, which shows us how long it took to render 99% of all frames.
In instances where there aren't major stutters, this should generally line up with average FPS. But that's the thing, average FPS completely misses situations where things are going wrong. It only polls data once a second, yet it could be averaging out 70ms frames with 8ms frames over a second to show you 80fps.
The more I'm researching this stuff, the more I can't believe that FPS has been a standard for so long. It's sooooooo bad.
I have med-small hands, and I definitely cannot claw grip my sensei. Spawn/Xornet is the way to go with claw, but as you are left handed, that's definitely not a good fit. I'll do some research for you and let you know if I find some good stuff.
FallingEdge
Member
So, built the new PC. No problem. Everything was fine. Booted up and installed Windows. Installed my programs and even went to overclock. But something was wrong. CPU was idling around 70 degrees C. When running Prime95, it was 102. Looking online, it said that an improper installation of the heatsink or poor application of the thermal paste can cause a substantial increase in temperature. Well, opened it up, took of heatsink and...
Yeah, forgot to take off the sticker. No metal to metal connection. No heat dissipation. Thermal paste did nothing.
However, after removing it, temps were idle around 38-40 with stress testing overclocked @ 4.3 was at 85. Will try for 4.5 later.
HUZZAH!
Yeah, forgot to take off the sticker. No metal to metal connection. No heat dissipation. Thermal paste did nothing.
However, after removing it, temps were idle around 38-40 with stress testing overclocked @ 4.3 was at 85. Will try for 4.5 later.
HUZZAH!
My man. I have the 500r and h100.
It was a pain getting that h100 installed so just know that going in. I have the 2 fans in push configuration. You could look into replacing the side panel fan with two 120 or 140 fans. If you do you might have to call corsair and have them send you the screws. My case currently houses a 2600k, 2x 580, and h100 and I've been pleased with it.
Aren't those temperatures still a bit high?
500R is a great case. Looking at that and the 800D, I kind of wonder why Corsair has so consistently put out overpriced garbage. 600T/650D has a bad design, 400R is way overpriced, 300R is outshined by cases in the $30-50 range, as is the 200R. The 550D doesn't hold a candle to the R4 or Ghost. The C70 while good, is $40 over where it should be (unless it is on sale).
Here's hoping the 900D at $350 delivers. That seems like a low price for what the case is offering. A comparable Little Devil, Caselabs, Silverstone, or Lian Li case are $500 or more.
I really fear that company becoming complacent like EVGA and Antec, where the quality slips because they can just bank on the name. But then they go and release something like the AF/SP fans that just destroy all of the competition.
Also, I'm quoting this for the new page because people need to see it.
Here's hoping the 900D at $350 delivers. That seems like a low price for what the case is offering. A comparable Little Devil, Caselabs, Silverstone, or Lian Li case are $500 or more.
I really fear that company becoming complacent like EVGA and Antec, where the quality slips because they can just bank on the name. But then they go and release something like the AF/SP fans that just destroy all of the competition.
Also, I'm quoting this for the new page because people need to see it.
That's because Min/Max/Average is a really bad way to measure performance. For example, here's 3 minutes of gameplay in Firefall. First chart is min/max/average, second chart is a plotline of the FPS.
Minimum 83 FPS, pretty good right? Wrong. Here's a plotline of the exact same data that shows, in milliseconds, how long it took to render each frame.
There are a huge number of frames that are taking over 16.7ms (60fps) to render, with some reaching 70ms (14fps). These are the kind of stutters that one would encounter when running out of VRAM.
A much more accurate measurement of 'average' experience is 99th Percentile Frame Latency, which shows us how long it took to render 99% of all frames.
In instances where there aren't major stutters, this should generally line up with average FPS. But that's the thing, average FPS completely misses situations where things are going wrong. It only polls data once a second, yet it could be averaging out 70ms frames with 8ms frames over a second to show you 80fps.
The more I'm researching this stuff, the more I can't believe that FPS has been a standard for so long. It's sooooooo bad.
FallingEdge
Member
I have a Hyper 212 EVO. Should it be lower? As mentioned it is OC to 4.3 (i7 3770K) but the VCORE is auto right now as opposed to being set manually. Don't know if that makes a difference.
Frequency matters very little compared to voltage when it comes to temperatures.
Turn off auto voltage, because I guarantee it is feeding your processor way too much.
FallingEdge
Member
Word.
http://www.overclock.net/t/1291703/ivy-bridge-overclocking-guide-asus-motherboards
I followed that guide and just left it at auto at first. It was getting late and I was tired.
I will set VCORE to 1.20 and try again tonight @ 4.3.
Thank you for the help.
sorijealut
Member
That is exactly the sort of stuff I was wondering about, when comparing 660 Ti and 7950 with average FPS.
Also, I can also confirm that cooler master mice are best claw grips I've ever used, period.
Curious about SS Sensei though...
Also, I can also confirm that cooler master mice are best claw grips I've ever used, period.
Curious about SS Sensei though...
Anton668
Member
FallingEdge
Member
So apologies for the noob question, but one of my friends was saying that OC day one on a new CPU is a bad idea. I should be letting it burn in a little at first. Thoughts?
DJ_Lae
Member
I've got one too, sitting waiting for the rest of my parts to arrive.
It seemed to review really well. Only complaint I've seen is that it can be a bit noisy.
My $0.02 from the Gaming mice thread. Keep in mind the mouse I otherwise use is the Xornet/Spawn. In fact, I have a Xornet here at work.
Bad fans, bad airflow design, fan controller that dies constantly, inconsistent tooling and paint.
I've gone through two personally, and have built in one for a friend. My second one I cut up to do what I wanted it to do, as I loved the aesthetics and some of the design philosophy. In the end, I just had to move to something that was a bit better.
Compared to the Switch 810 and Shinobi XL, which are the same price, it's just not very good. Heck, it's not very good compared to the Scout, 500R, Arc Midi, Define R4, yada yada.
MarshMellow96
Member
How does the CM 690ii Advanced do in your eyes?
sorijealut
Member
Same news here as well. I own MX518, G400 and various other logitech mice, but I can never click faster than Xornet. These switches on this mouse are so good.
Since you own both mice, that speaks volumes for Xornet... I do imagine Sensei to be a bit restrictive(?) with that form factor.
It's awesome. Hard to beat still.
CoolerMaster is pretty tough to beat in terms of price:quality. They've been at it for so long, and consistently try reinventing themselves. Even the 690 has gone through four iterations in as many years just so they can keep the design fresh with the newest and greatest ideas. Not many companies out there do this sort of thing. In fact, Fractal and CM are the only ones that come to mind.
CoolerMaster is pretty tough to beat in terms of price:quality. They've been at it for so long, and consistently try reinventing themselves. Even the 690 has gone through four iterations in as many years just so they can keep the design fresh with the newest and greatest ideas. Not many companies out there do this sort of thing. In fact, Fractal and CM are the only ones that come to mind.
If I didn't have 1000 other things to buy, I'd replace the Xornet with another Sensei
sorijealut
Member
If I didn't have 1000 other things to buy, I'd replace the Xornet with another Sensei
LOL oh, you confuse me...
darn you.
This is the single lowest price I've seen for the full model. MONEY MONEY MONEY I'VE NO MONEY
doniewhalberg
Member
I did ask in headphone GAF, but since that may be a music orientated thread, rather than gaming, I figured I'd ask here too. Apology if it's an unwarranted double up.
Anyway, got my m-itx system pretty much figured out, and today I've been looking at sound solutions given I can't get a soundcard in there. I read up on good headphones, DACs and so on, and from what I've gathered, i guess if I just went ahead and bought something right now, I'd go for a pair of Ultrasone Pro550's and the FiiO E17 DAC. Is this even remotely in the right ballpark for someone who will be very much primarily playing games. This won't be for music use or anything really. If this sounds horrible/outdated or anything, please let me know haha. I won't be purchasing until I've did the pc in March, but just trying to nail down what I want now so I can budget accordingly.
Anyway, got my m-itx system pretty much figured out, and today I've been looking at sound solutions given I can't get a soundcard in there. I read up on good headphones, DACs and so on, and from what I've gathered, i guess if I just went ahead and bought something right now, I'd go for a pair of Ultrasone Pro550's and the FiiO E17 DAC. Is this even remotely in the right ballpark for someone who will be very much primarily playing games. This won't be for music use or anything really. If this sounds horrible/outdated or anything, please let me know haha. I won't be purchasing until I've did the pc in March, but just trying to nail down what I want now so I can budget accordingly.
Meh, it's fine. I'd say an EXTREME OC with hours of stress testing and benches can be quite rough on a processor. It's not like a car engine where things are actually still sealing during a break in period though. I'm not an engineer so maybe there's something I don't know, but, this sounds like one of those 'dude who equates all PC stuff to autos once said it on a forum and used analogies to prove his point so people bought it and repeated it a ton' type situations.
Trojita
Rapid Response Threadmaker
Yeah, but you can just record your gameplay or just pay close attention.
If you constantly go past your target while tracking or snapping, reduce the corresponding X/Y CPI. If you aren't able to keep up or fall short, increase corresponding X/Y CPI.
If you are tracking well but fail on snapping, then increase/decrease mouse acceleration.
I'm right at about 900/700 on my X/Y for modern shooters, and 900/900 for proper 3D FPS games like Warsow/T:A.
If you constantly go past your target while tracking or snapping, reduce the corresponding X/Y CPI. If you aren't able to keep up or fall short, increase corresponding X/Y CPI.
If you are tracking well but fail on snapping, then increase/decrease mouse acceleration.
I'm right at about 900/700 on my X/Y for modern shooters, and 900/900 for proper 3D FPS games like Warsow/T:A.
This can't be true at all.
The 550D is built like an absolute beast, and has all the amenities for silent performance and fantastic airflow. I've never used the R4 to say on it, but the 550D scores as high or higher in all the reviews that I've seen.
It's also an average of 30$-40$ more than the R4. That's the only negative, imo. I happen to catch a deal that brought it down to R4 levels, though, so I'm good lol.
That's the core of the issue.
Build quality is a bit lower than the R4, from what I've encountered. The paint isn't nearly as good either. There's a lot of really chintzy little things on the 550D, which is basically a gussied up 300R. But the kicker is the price. Ghost is generally found for $80, R4 for $100, and 550D for $130-150. That increase in price is just not worth it at all.
The Ghost has some pretty cool features that aren't found elsewhere, and is incredibly light while being well-built. There's no shortcuts like washers glued directly on cut mesh that you find on the 550D.
For the savings, you can buy either the Ghost or the R4, get 3-4 really nice fans, and *still* be below the cost of the 550D.
I admit, I'm extremely picky when it comes to cases. I go through tons and build often for other folks, so it's the thing that I'm exposed to more than any other component. I also place a greater importance on it than other people do for a variety of reasons.
That's because Min/Max/Average is a really bad way to measure performance. For example, here's 3 minutes of gameplay in Firefall. First chart is min/max/average, second chart is a plotline of the FPS.
Minimum 83 FPS, pretty good right? Wrong. Here's a plotline of the exact same data that shows, in milliseconds, how long it took to render each frame.
There are a huge number of frames that are taking over 16.7ms (60fps) to render, with some reaching 70ms (14fps). These are the kind of stutters that one would encounter when running out of VRAM.
A much more accurate measurement of 'average' experience is 99th Percentile Frame Latency, which shows us how long it took to render 99% of all frames.
In instances where there aren't major stutters, this should generally line up with average FPS. But that's the thing, average FPS completely misses situations where things are going wrong. It only polls data once a second, yet it could be averaging out 70ms frames with 8ms frames over a second to show you 80fps.
The more I'm researching this stuff, the more I can't believe that FPS has been a standard for so long. It's sooooooo bad.
I have med-small hands, and I definitely cannot claw grip my sensei. Spawn/Xornet is the way to go with claw, but as you are left handed, that's definitely not a good fit. I'll do some research for you and let you know if I find some good stuff.
I love you.
Keep up the good work.
Decided to get rid of my old CPU and I'm cross posting this from the buy/sell thread.
After upgrading my PC I've decided to put my old stuff on sale. All prices shipped.
i5-750 - Used for a couple of years now and still works fine $95 (Intel heatsink never used BTW)
ASUS P7P55D-MB Same as above two years now $45 (comes with what you see in the pic)
If buying together then $130.
Corsair 500R case (comes with all fans and screws) $95 (local pickup only. If you live in the Chicago suburbs of Skokie, Niles or Morton Grove or if you are willing to drive up there)
It has a broken USB 3.0 to USB 2.0 cable thing. It still works though (I bent the pins back myself) Some light scratches as well, it was bought and used for a few months last year.
Koroviev
Member
That's because Min/Max/Average is a really bad way to measure performance. For example, here's 3 minutes of gameplay in Firefall. First chart is min/max/average, second chart is a plotline of the FPS.
Minimum 83 FPS, pretty good right? Wrong. Here's a plotline of the exact same data that shows, in milliseconds, how long it took to render each frame.
There are a huge number of frames that are taking over 16.7ms (60fps) to render, with some reaching 70ms (14fps). These are the kind of stutters that one would encounter when running out of VRAM.
A much more accurate measurement of 'average' experience is 99th Percentile Frame Latency, which shows us how long it took to render 99% of all frames.
In instances where there aren't major stutters, this should generally line up with average FPS. But that's the thing, average FPS completely misses situations where things are going wrong. It only polls data once a second, yet it could be averaging out 70ms frames with 8ms frames over a second to show you 80fps.
The more I'm researching this stuff, the more I can't believe that FPS has been a standard for so long. It's sooooooo bad.
I have med-small hands, and I definitely cannot claw grip my sensei. Spawn/Xornet is the way to go with claw, but as you are left handed, that's definitely not a good fit. I'll do some research for you and let you know if I find some good stuff.
Oh, cool. This is the sort of data I was looking for.
- Status