EPTSimple Questions Simple Answers - Page 310
| Forum Index > Tech Support |
|
Az0r_au
Australia385 Posts
| ||
|
Cyro
United Kingdom20327 Posts
| ||
|
xasuma
Chile62 Posts
Question here: - What are the practical differences between Rams, of 1600 mhz and lets say 2400 mhz?? Do they affect gaming experience much? What does the speed different really makes a difference in (where would you see the biggest improvements). Thanks, I just wanted to know since I am considering a different motherboard that supports higher ram speeds in the near future. | ||
|
Myrmidon
United States9452 Posts
A lot of times, on higher graphics settings, it's the graphics subsystem holding things up. If not, then faster RAM could help a bit, but that usually doesn't happen in most games unless you're looking at differences at much higher than 60 fps, which is mostly useless anyway. http://www.xbitlabs.com/articles/memory/display/sandy-bridge-ddr3_7.html#sect0 http://www.anandtech.com/show/4503/sandy-bridge-memory-scaling-choosing-the-best-ddr3/6 Generally the difference is very modest, and 1600 MHz is plenty, considering the cost. You should be worrying about CPU clock rates before RAM frequency. If you're looking at say an overclocked i5-3570k and really play certain demanding games and expect the best, you could shoot higher than 1600 MHz maybe. That said, certain kits at a small premium over the lowest-priced 1333 / 1600 MHz kits may be able to run at very high clocks, so even then it wouldn't be worth getting the expensive RAM. | ||
|
hacklebeast
United States5090 Posts
1. Is there a way to make chrome interact with flash better? 2. It works fine once it the error message "shock wave flash is unresponsive" and I am able to kill it, restarting flash. Is there a faster way to kill flash rather than wait for the message to appear? | ||
|
ZeratuLsc2
Canada426 Posts
| ||
|
Rannasha
Netherlands2398 Posts
On April 13 2013 17:12 Hellfury wrote: What kind of upload speeds are required for streaming? That completely depends on the resolution/quality you want to stream with. You can make do with 500 Kbps, but if you want a supercrisp 1080p stream, count on 5 Mbps or more | ||
|
Blaec
Australia4289 Posts
On April 13 2013 16:24 hacklebeast wrote: Once every hour or two, chrome starts vomiting all over itself and becomes completely unusable due to it's incompatibility with flash. I've had this problem on 3 machines at this point, so I assume this is not isolated to me. 1. Is there a way to make chrome interact with flash better? 2. It works fine once it the error message "shock wave flash is unresponsive" and I am able to kill it, restarting flash. Is there a faster way to kill flash rather than wait for the message to appear? Flash is shit, dunno how to fix your issues. But you can kill flash quicker by having resource monitor or task manager open and using either to kill the FlashPlayerPlugin process. | ||
|
Ropid
Germany3557 Posts
On April 13 2013 16:24 hacklebeast wrote: Once every hour or two, chrome starts vomiting all over itself and becomes completely unusable due to it's incompatibility with flash. I've had this problem on 3 machines at this point, so I assume this is not isolated to me. 1. Is there a way to make chrome interact with flash better? 2. It works fine once it the error message "shock wave flash is unresponsive" and I am able to kill it, restarting flash. Is there a faster way to kill flash rather than wait for the message to appear? Chrome comes with its own Flash packaged. You can force it to instead use the normal Flash that's manually installed, sitting somewhere in the Windows directory and used by other browsers like Firefox. To do this, go to "chrome://plugins" in the Chrome location bar. Make sure to click on "+ Details" somewhere in the top right of that page. Disable the Flash plugin that's not in the Windows directory. Reload pages that use Flash. | ||
|
xasuma
Chile62 Posts
Now since I feel like learning a little more, another simple question has to do with overclocking voltage. - I get the basic idea of overclocking and making sure that the system or the specific component doesnt get too hot and what not. But I am clueless when it comes to the voltage adjustments. I will put my CPU intel i5 2550k as an example here. I overclocked it using AXTU (the asrock utility extreme tuning that came with my motherboard: "Asrock p67 pro3"). And I just up'd the cpu ratio to x42 so i ended up with a 4200 mhz speed. All the voltages were left intact. (Since I didnt know what to do with them) . I have had this configurations running for about 2-3 weeks and it is stable. Here are some pics of all the voltages and the settings. + Show Spoiler + ![[image loading]](http://oi49.tinypic.com/15vaxd.jpg) ![[image loading]](http://oi48.tinypic.com/2h6aomh.jpg) Those are idle temperatures. And I just want to know what voltage should I adjust, when and how much? So in the future I know exactly what I am doing. | ||
|
Ropid
Germany3557 Posts
There are two methods to change the voltage when overclocking on your CPU and board. You can change an offset. That offset will be added to the VID the CPU demands. With this, you can tweak the default behavior to get the CPU stable at higher MHz where Intel's default VID calculations are off. With this method, the CPU voltage will still go down when idle. The other method is a complete override of the VID behavior. If your BIOS has that option, you can set a fixed voltage. If it doesn't, you can still simulate that behavior by disabling all power saving features of the CPU (EIST and C1E). If you do that, your CPU will always be at 4200 MHz and always want the same voltage. You then use the offset setting to change the CPU's voltage. There's another strange behavior about idle and load voltage. The "load line calibration" (LLC) setting changes how aggressive the board is in trying to keep the load voltage at a constant level. This is off by default, and can cause a headache trying to get your PC stable when overclocking. You have to try the different settings (50 % should be good) to find which one keeps the voltage most stable. Using the offset method complicates this. EDIT: ASRock overclocking guide: http://www.overclock.net/t/1198504/complete-overclocking-guide-sandy-bridge-ivy-bridge-asrock-edition There's a zip with a bunch of useful programs in that guide. Don't look at the ASRock tool for temperatures, for example. The CPU itself has four sensors for its four cores, which will show much higher temperatures than the motherboard sensor you'll get to see in the ASRock tool. The zip also has simple programs that put a lot of load on the CPU to see what would happen to temperatures in the worst case possible. | ||
|
ghindo
United States58 Posts
| ||
|
sc_a.M
420 Posts
Is this working as intended, that a language change results in a cleared account? The thing that bothers me the most is that my friend list is empty. It's kinda pissin me off tbh. Does someone have knowledge over said problem? I don't consider this in feature =/ | ||
|
Rachnar
France1526 Posts
| ||
|
thom
England105 Posts
Make sure you're on the correct region! | ||
|
sc_a.M
420 Posts
On April 14 2013 04:23 thom wrote: On the login screen, on the left hand side you can choose which region you want to log into. Make sure you're on the correct region! ok gratitude :D It's strange it only lets you chose region when you unable the password field. Thanks very much for the help (: | ||
|
mav451
United States1596 Posts
On April 14 2013 03:30 ghindo wrote: Is there a good site for doing direct benchmarks between two CPUs/GPUs? I've been using Anandtech for a while, but its database seems to be missing a few of the parts I want to look up. I would say it is better to search for this information on a case-to-case basis. For a quick reference in a pinch, Anandtech may be ok, and is certainly some notches abvoe bottom feeders like Passmark...but that's it. Would you mind elaborating what you're looking for specifically in terms of hardware? What game or software application are you looking into? GPU drivers and game patches can significantly change what is considered the better 'value' card, and I highly doubt Anandtech does any diligence on that front. So again, your best results would be finding the absolute latest information using the latest drivers. You generally never want to rely on these generalized benchmarks in making your purchasing decision. | ||
|
Belial88
United States5217 Posts
The number of good review sites out there I could count on one hand and you've probably never heard of them, and even good sites do bad reviews. Tomshardware, Anandtech, generally do really, really really bad reviews (TH's reviews on thermal paste and heatsinks isn't the worst on stock fan configurations, but heatsink performance on stock fan configuration is pretty worthless). Your best bet is just ask on a knowledgeable forum about the hardware in question (not TH, not anandtech forums). Overclock.net, overclockers.com, ocuk, xtremesystems, that's really it ([h] has a great site but a terrible forum). Note that a lot of forums, like overclock.net, actually have reviews or people who do reviews, that post great articles, and unlike TH/Anandtech/etc, the reviewer will (have to) field criticism and comments from people, so you'll get more in-depth information, see what they did right and wrong. The problem with all the big sites is they are paid to do reviews. The big problem with that, besides obvious ethical quandries, is that the companies force them to cover certain subjects. So you'll just only hear about how awesome and great some motherboard's Virtu MVP is, and how amazing the software is, when in reality it's nearly useless, no one uses it, and the very few that do use it, only gain a small benefit in a niche field. They'll also be forced to review things that are totally useless like a motherboard's included overclocking program, something that overclockers don't use (software for cpu/ram overclocks), and can often be used on differing brands of boards (ie Asus' memtweakit can be used on gigabyte boards, etc). So then they put this image out there that one board is better than another because of how the software on it is, despite the fact the software has nothign to do with the hardware and that the board might actually be crap. You'll learn more and more about this, asking why aren't certain questions being answered on review sites, but sites like tomshardware and anandtech are downright misleading and dangerously wrong. It's very hard to tell if sites like, say, anandtech are either incompetent or just intentionally aware of what they are doing but want to appease the big companies and want more followers (and the dumber your articles, the more people will read them, sadly). For example, they compared Z77 motherboards ability to overclock by taking software voltage readings in HWMonitor, but in reality HWMonitor was bugged with Gigabyte (which was well known but AT didn't get the memo) so they said gigabyte wasn't as good, and then they said asrock was the best when in reality the firmware in asrock boards calculates voltage over-generously and way underreports the true voltage, and the hardware on the asrock board had the worst quality of all the boards (it's like saying a civic is a way better car than a lamborghini because the speedometer has a bigger range on the civic). Then they commented on factors like ripple and voltage noise when you simply cannot see those using software, which everyone knows but apparently AT either didn't or chose to lie about anyways. xbitlabs, vortez, hardocp, they have some good reviews, but honestly I'm not even sure how good they are because I simply don't rely on review sites anymore, they are just the ones that I looked at and didn't see glaring issues with. Tomshardware (okay tests on stock fan configs on heatsinks but that's a useless test), anandtech, benchmarkreviews, hardwaresecrets (good explanations of things, bad reviews and how-tos)... more bad than good. sin0822, johnnyguru, they have some awesome reviews but they are both literally just forum posters who created a website (which is why they are so good) and aren't too prolific (or active). | ||
|
Myrmidon
United States9452 Posts
btw the big sites don't get paid, unless you count via traditional advertising like product links / banner ads and the like or the fact that often they get to keep review samples. Also, review content is up to their own discretion, not at all enforced—in the rare case that a company tries to play games like AMD did with staggered NDA dates on coverage material for desktop Trinity, all the major review sites were bitching about it. It's mostly just laziness, lack of time, lack of money, and incompetence that is the problem usually, not ulterior motives or monetary considerations. It took how many years for people to do capture-based or even high-speed-camera-based analysis of graphics performance? Nobody does thermal imaging for case reviews. Almost everybody does noise measurements with a high noises floor or poor equipment. People rarely do any hardware voltage or temperature measurements. Audio benchmarks are a joke. It goes on. A great number of tech writers for review sites don't have technical backgrounds. The ones that do, generally don't have backgrounds in relevant areas of electrical or computer engineering, computing, etc. And the few that do, don't actually have relevant experience in the industry doing the actual R&D, QA, etc. That said, the analysis of the small guys may often be suspect too. | ||
|
xasuma
Chile62 Posts
| ||
| ||
