nekro

Ryzen 7 1700 if i was you! more core more threads will ne more usefull to use. But maybe a cheaper 1600X according to your needs could be a better option.  Everything is based upon what uses you will have with the cpu and your budjet. If you have the money go grab the 1700!

The highest stable clocks you can manage, without artifacts or instability, without increasing the volts would be pretty safe.
 
Any crashes or lock ups during the oc'ing process don't harm the gpu, and just server as a reminder that you have gone past its limit, but pumping in the volts, even if the temps are not too bad can degrade the life of the chip, and the other components on the card like the vrm's.
 
Considering you have the g1 gaming version of the card, i highly doubt you will reach anywhere at/ above 80C, so that will make finding the max oc a little easier. As previously said just increase the core by increments of around 20 mhz, and run stress tests for a bit, to see that nothing weird is happening. Rinse and repeat until you find where it starts to crash, and then back of by 20 mhz, and you have you max stable core clocks (probably around the 1500 mhz in-game clocks). 
 
Then you can look at the memory, where you can increase it by larger increments, as the memory dosent contribute as much as the core does when overclocking, and fine tuning it isn't necessary.

I know exactly what the Sandy jump was, and Haswell is an equal jump if your code is properly parallelized. The problem is all the software comparing Sandy and Haswell is still not using the parallel instruction sets which arrived for Sandy. That 4th ALU is a huge performance boost if you compile code which can use it.If we compiled all existing benchmarks using the Intel C++ compiler (and inserted a fair dispatcher) you'd see Haswell kicking the snot out of Sandy Bridge.

Broadwell is and isn't a dramatic improvement. With TSX-NI to work in upper SKUs, we'll see some Multithreading improvement over Haswell. Also, floating point calculations take 40% less time due to the shrink shaving the FPU instruction time from 5 cycles to 3.
But in integer where almost everything was already 1 cycle per instruction even for vectors, you're not going to see big IPC increases anymore without more advanced compilers and even wider instructions (like 512-bit AVX for Skylake Xeons).
Serial processing is reaching its limits. Instruction level parallelism already has reached its limits. We need to dive into data-level parallelism and just write better code.

Once you establish that kingdom you'll better understand the game's mechanics having gone through probably a few rulers. Also, it means you Start out with primogeniture inheritance which keeps things together nicely. If you need any help or have started elsewhere feel free to message me for advice/tips. I know how daunting it can be.
Random shite my phone keyboard won't delete: r starrwo

CK2 has a really steep learning curve but it is extremely rewarding. I've sunk a good couple hundred hours into it. For a first play through I'd suggest disabling a lot of the DLC, especially the Viking one if you plan on playing in northern Europe/Britain/Ireland, as the vikings can be extremely challenging early game.
I'd secondly suggest starting in Ireland for the first time because Ireland is not part of a kingdom, your enemies are not going to overwhelm you, you can build up duchies and eventually the kingdom of Ireland itself. Once that's completed pull oull OnceOnc
Sorry I'll continue this in a new post my phone's keyboard is fucking uup

Crusader Kings 2 is great and eats the hours up pretty quickly. It's a strategy game though so it might not appeal to you, but it's great because you can play any nation in any part of Europe, North Africa and some of Asia (so obvious examples are Byzantium, the HRE, England, France etc). It's very deep and there are a lot of things to manage such as factions and plots, uprisings, threat of invasion by a neighbour etc. Easily one of my favourite games, though I've never really enjoyed studying the Middle Ages myself, I find it to be very dull. Ancient history is where it's at.

Awesome!

I'm assuming you mean you'll use dxtory's DirectShow output to send video to OBS to capture, in which case you are right to think its a bad workload. It won't have any performance increase or decrease since running Dxtory and capping with OBS is LITERALLY the same as using OBS's game capture source, with the added value of having to run dxtory's program in the background also. I would recommend Dxtory for recording if you use it with lagarith already because you can't record with custom codecs on OBS, maybe something they'll implement? Who knows

The 1000Kbit/s bitrate used for offline recording is just a guideline. OBS has some weird issues (not really issues but more like behaviours) when it comes to localized recordings, it doesn't really like to stick to the bitrate that you set when you aren't using a constant bitrate, so the bitrate is going to far surpass the 1000Kbitrate. Also, as you said, anything above 128 for the audio bitrate is fine for localized recording, and the sweetspot is around 192kbps, but youtube can also do much higher bitrates when using ~4k resolutions (up to like 512Kbps)
 
EDIT: Check HERE for more info on the YouTube video and audio encoding settings I'm talking about (if you didn't already know)
 
Also, I should mention that recording in flv or mp4 is purely objectional. It is known that flv encodes faster and has lower file sizes (while maintaining the same overall quality) but  flv isn't widely accepted by most video editing programs,  so mp4 is likely the way to go if the certain conditions are met, such as editing, enough CPU bandwidth to encode, and enough hard drive space.
 
I appreciate all the interest in this thread so thanks for your comment 

First of all, Sony =/= SOE. SOE was just a small sector of their gaming division. The least profitble by far.
 
Second, Sony isn't dying any time soon. http://i.imgur.com/YsJmzsH.jpg
 
This myth does need to die, though.

There hasn't been a proper update to PS2 in ages. I lost interest a long time ago mid last year after putting many hours into the game. I could probably be considered a vet on the ground. There are many reasons why the game made me lose interest but for the most part it's the performance issues and the game just didn't take it far enough. When I say not far enough I mean they simplified the entire games mechanics too far in terms of vehicles. What I personally would have liked to see was War Thunder level of vehicle fights but with soldiered on the ground as well on the huge scale of Planetside. Maybe the next game will be better, but I doubt they will ever make it. Also why did they ever make it Free 2 Play, it could have maybe worked better if it was £100 for the game and you get everything. Leveling up in games is just an artificial way of keeping the player interested but that isn't fun i the least.

SOE claims that the job cuts will not harm the future of Planetside 2 and other running games. I find this unlikely, the game at the moment is an unoptimized badly designed mess ( I am talking about the engine here). It hasn't changed much ever since the game came out. With less people working on the game I doubt we'll ever get that game to use more than 1 core properly so we can get out CPU's past 20% CPU usage which is ridiculous. People pay hard earned money buying a really expensive CPu and then when they want to play their game they get very similar performance to someone who spent a fraction of what you did. My GPU for example in on relax mode when I play that game idling about 30% usage and same with my CPU. If they game used 90% of both my CPU and GPU like in properly designed 64bit games I get 120FPS easily. They should have never designed the game for a 32 bit 1 core cpu computer because its bitten them in the ass. And now we hear about lay offs this is terrible, how exactly are they gonna get this game to run properly. 
 
I imagine that SOE is banking on the console market for Planetside 2 which could improve performance on PC to however I find this unlikely. Console gamers stick with the same games they play all year round, that being either BF3/4 or COD. They will doubtfully ever play PS2 let alone pay money on that platform so they are effectively letting PS2 on PC to rot. I do know they will be releasing content but it would seem every new patch they release it ruins the performance even more or breaks that game until they release a couple new hotfixes to bring the performance back and that is with a full team, if PS2 has 100 less people working on the game how exactly will they ever patch the game in a reasonable amount of time. I know I this may sound very ranty and well it is simply put I think this was a terrible idea to launch the game in the first place if it wasn't designed properly top begin with and they to jump ships like this effectively just pisses me and many other PS2 players off, I would imagine. 

Great games? 

You mean butchering Star Wars: Galaxies? Pushing PS2 to be more and more Free 2 Pay? 

Good goddamn riddance. 

Whats your total budget? and does that include os/monitor/peripherals?
 
Heres what I'd be looking at if you're on a very tight budget:
An R9 280 is a better choice than the 960 as the performance is very similar, but games are starting to need more than 2gb of VRAM, thus making the R9 280 a better choice. Plus an R9 280 is about £30 cheaper than a 960. 
PCPartPicker part list / Price breakdown by merchant   CPU: Intel Core i5-4440 3.1GHz Quad-Core Processor  (£131.20 @ Aria PC)  Motherboard: MSI H81M-P33 Micro ATX LGA1150 Motherboard  (£33.01 @ Aria PC)  Memory: Kingston Fury Red Series 8GB (1 x 8GB) DDR3-1866 Memory  (£51.46 @ Amazon UK)  Storage: Seagate Barracuda 1TB 3.5" 7200RPM Internal Hard Drive  (£35.94 @ Aria PC)  Video Card: Gigabyte Radeon R9 280 3GB WINDFORCE Video Card  (£137.99 @ Amazon UK)  Case: NZXT Source 210 (Black) ATX Mid Tower Case  (£32.99 @ Novatech)  Power Supply: EVGA 500W 80+ Bronze Certified ATX Power Supply  (£38.52 @ Amazon UK)  Operating System: Microsoft Windows 8.1 (OEM) (64-bit)  (£72.35 @ Aria PC)  Total: £533.46 Prices include shipping, taxes, and discounts when available Generated by PCPartPicker 2015-01-31 17:17 GMT+0000

"I'm proud of my son, Linus" - Richard Branson

Problem "solved" for me.
 
Had made the order for an Asus GTX 970 STRIX yesterday and today called the store and talked about this problem. 
 
They gave me a 20% off in an Asus GTX 980 STRIX.  
 
So I got that going for me, which is nice. 

Source via PCPer
 
http://www.anandtech.com/show/8935/geforce-gtx-970-correcting-the-specs-exploring-memory-allocation/4
It appears this is a classic case of false advertising. Nvidia's marketing team has been known to stretch the truth with their marketing in the past. For instance claiming a 165W TDP for the GTX 980 (which is false) to claim a 2X perf/watt boost over Kepler.

 
Another example is advertising the Tegra X1 chip as a 1TFLOP "supercomputer" when in fact they're using half precision peak FP16 performance figures. A performance metric which was abandoned from Shader model 3.0 onwards and utterly unsuitable for a "supercomputer" which not only requires FP32 but even more so double precision FP64 compute. In which the Tegra X1 can only deliver less than 2% of the advertised compute performance.

make a strawpoll
They should already have done something about it. With the real specs out this card is like a 965 or something. Ramgate!

EDIT: Someone that works for Nvidia responded in the geforce forums (link: https://forums.geforce.com/default/topic/803518/geforce-900-series/gtx-970-3-5gb-vram-issue/post/4438090/#4438090) Thanks @Templar848 for the link
So it seems that they're proud of the card and sorry for the issue. He said that they're also working on a driver to improve memory allocation (I thought this was by design and not by drivers/optimization because Nvidia said that)
If you really don't want the card, it seems that you can ask for a refund to the store where you got it, if they don't want to give you that you can ask him for help
Just wanted to know your opinion guys.
I paid more than $380 USD for a "64 ROPS, 2048 L2 cache" device.
Now Nvidia is saying "oops, sorry, it's actually 56ROPS and 1792KB of cache".
(Article: http://www.pcper.com/reviews/Graphics-Cards/NVIDIA-Discloses-Full-Memory-Structure-and-Limitations-GTX-970)
What you guys think
I find it similar to the Intel Pentium stuff where they lied to customers.

Just take it for a drive
 
On topic: http://thehowtoman.com/how-to-unpark-cpu-for-battlefield-4/

Terve!
I'll throw in my stuff as well, because I made other experiances.
 
 
You can be right here. I never tested two dual channel kits in quad channel (mainly because I do not have a mainboard to support it).
Anyway: If you want to run quad channel, please use a quad channel kit. There is nothing more frustrating than RAM that does not accept each other and as soon as you are in the quad channel pricerange a few more bucks do not count anymore.
 
 
That is not true, at least from what I have experianced. If you have a quad channel kit in a dual channel board it will simply run in dual channel with its bandwith. The RAM cannot force the mainboard to use quad channel.
Both of my machines that have that kind of RAM configuration rund fine. At least in terms of RAM.
 
 
 
So, to answer your question at last:
Yes, it is absolutely possible and you can install it as if it was a dual channel kit (I guess you know that?).
Channel means, as said, "a link where RAM and CPU can communicate with each other".
I hope I clarified all the thing in here.
Sayonara!

As indicated below, the specs and 3dmark 11 score




The GeForce GTX 960 scores P9960 and X3321 points in 3DMark 11, 6636 points in 3DMark Fire Strike, 3438 points in 3DMark Fire Strike Extreme as well as 1087 points in 3DMark Fire Strike Ultra, according to PCEva.com.cn. If the numbers are correct, then the GeForce GTX 960 will offer performance – at least in 3DMark benchmarks – comparable to that of the GeForce GTX 680 and the GeForce GTX 770.
 
This should put it between the 770-780 range of performance while keeping at a comparable price with the 280/285 (~210$US market price).

Tell me what you think about the formatting of the topic, and post what you're thoughts are below.

http://wccftech.com/nvidia-geforce-gtx-960-reference-overclocked-performance-revealed-performs-slightly-faster-radeon-r9-280/
http://www.kitguru.net/components/graphic-cards/anton-shilov/specs-and-performance-of-nvidia-geforce-gtx-960-revealed/
http://videocardz.com/54263/nvidia-geforce-gtx-960-specifications-and-performance-leaked

If this is a repost. One person say it is in a kind matter. I honestly spend 5 minutes looking for gtx 960 and nvidia from method Site:linustechtips from both past 24 hours and the past week for anything of the same topic.
 

Can you be a bit more specific? What is 'this' and damping of what?
 
Edit: talked to SSL via PM. Maybe this explanation is interesting for others too:
All drivers have their own damping, caused by the spider and surround pulling the cone back to its original position. The air compliance in an enclosure helps the driver to return to its original position and thus reducing excursion. Above the Helmholtz freq the air inside the enclosure is bouncing against the air trapped inside the port so the backside of the driver gets to use the air inside the enclosure as a spring. At a certain freq the air starts to resonate causing the output seen in the graphs of OP. The backpressure increases, limiting driver excursion severely, all output comes from the port. To illustrate I made a quick simulation of ported vs sealed. 

X-axis is freq, Y-axis is excursion. The dip is the resonance freq, under resonance freq the port is just a hole (no damping) causing excursion the increase rapidly.

Isn't it the effect that happens when you blow on a glass bottle of beer? . very nice analysis