Malefic Rage

Referring specifically to DCS: I'm using a Radeon R9 290 (slightly slower than a GTX 970) and get 60 to 150 fps at 1920x1200 resolution (16:10) at reasonable levels of detail. Unless moving to either 4k or VR, I'm not sure if you will get much value for money given that the 970 is faster than any card out when DCS was released. In general, I'm interested in the 1070 but will wait for AMD's new Polaris architecture to be released so I can both see what it has to offer and take advantage of the price drops. Interestingly the new NVIDIA architecture does not implement the DX12 'killer' feature: Asynchronous Compute whereas AMD's Polaris will. (Interestingly most of the current generation of AMD GPUs already do that). I think NVIDIA assumed that since Windows 10/DX12 isn't ubiquitous enough yet for developers to rely on, they can get by without implementing it in hardware yet.

Being a user of many different products including AMD/Intel/Nvidia I'm not sure where you get that idea from. AMD graphics cards are pretty much quality/performance/price on par with their NVIDIA counterparts. Their new Polaris architecture is a completely new design intended to compete with NVIDIA's newest products at the top tier of performance.

Also, with respect to GPU, I'd suggest waiting for the new NVIDIA GTX 1070. They will handle 4K DCS EDGE with ease. If you still want the top end, wait a little while for the GTX 1080 to go down in price; they are currently only selling a high premium "Founder's Edition". Cheaper ones will follow soon.

I also strongly recommend that you stay away from the socket 2011 i7 CPUs. These are expensive enthusiast parts with very high power and cooling requirements in addition to needing a socket 2011 motherboard. These usually require a liquid cooling solution. Due to the way DCS is programmed, the extra cores will give you no benefit whatsoever. If you want to buy the best CPU for DCS get an Intel Core i7 6700K. Even if you don't overclock, the K (unlocked) version gives you an extra 600 MHz over the standard i7 6700. Stick with the 1151 socket.

Here's one for the IT crowd: I read an article on Slashdot a week ago which explained how the US Air Force is planning on fitting high powered lasers on fighter jets. Some guy quips "Does that mean they'll be called HP Laser Jets?" Badum!

IMO the benefits of 1ms redraw rates are highly overrated. I doubt they've even been subjectively tested. Dividing 1 second (1000ms) by 1ms gives 1000 redraws per second. When you consider that the average monitor only refreshes at 60 to 144 Hz, there would be little benefit in theory. Even at a more common 5ms redraw rates you still get 200 per second. IMO get the better quality monitor. As a point of consideration though, the advertised redraw rates are usually in ideal testing situations. IRL you'll see up to 15ms in practice depending on the colour change.

Flight sims to tend to require more financial investment that typical games, especially if you really intend to get into them in depth. Between the need for a decent rig, the peripherals you buy and the modular nature of the product expansions, you'll end up paying a bit. Mostly because it is a relatively niche market. IMO the fun and satisfaction is worth it though.

Hello. Based on benchmarks from hardware review sites, latency is a rather tiny factor in memory performance compared to frequency. Also, going too far above the JEDEC standards (2133) gives diminishing returns since the CPU architecture is developed to work optimally with the official specs rather than the overclocked ones. Most of the improvements you get are in theoretical tests rather than real-world performance. Also, IMO 16 GB is quite sufficient for DCS even for large missions.

Thank you very much for that link. I hadn't come across that before. As for frequency, I should have mention that I tested the AMD CPU with an overclock of 4.2 Ghz. I got about an extra 5 MinFPS but it wasn't that significant. I believe the issue lies with the AMD architecture itself. (I noticed on that link all tests were using an Intel CPU). It will be very interesting to see their new generation coming out later this year. It is supposed to be a brand new architecture focused on per-core performance.

Extent of CPU bottleneck Hello. I know this has been discussed a few times, I just wanted to share the results of some experimentation. Until recently I had an AMD FX-6300 CPU with a Radeon R9 290 and have consistently had issues with FPS. Particularly significant was choppiness and minimum FPS when using TrackIR. Since my video card is relatively high-end, especially for the time DCS was released, and I only use medium settings, I couldn't understand the issue. After my motherboard died I replaced my CPU with an Intel Core i5 6600K which I overclocked from 3.5 Ghz to 4.2 Ghz. After running a test run on the same settings with the same video card and with no AI units my results are as follows. (I also tested with my AMD CPU overclocked to 4.2Ghz. I can't remember the exact results so I've listed approximations based on what I remember. AMD FX-6300 (3.5 GHz, 6 core) MinFPS - 21, MaxFPS - 70, AvgFPS - 38 AMD FX-6300 (4.2 GHz, 6 core) MinFPS - 26, MaxFPS - 80, AvgFPS - 43 Intel Core i5 6600K (4.2 Ghz, 4 core) MinFPS - 60, MaxFPS 120, AvgFPS - 90 The difference is spectacular. Basically it means that DCS will not play smoothly on any current-generation AMD CPU unless very modest graphics settings are used (particularly view distance) and only small missions are played. After observing the rendering stats, there seems to be a direct correlation between number of objects rendered and FPS. In hindsight I should also have taken results during a large mission but since my old AM3+ motherboard is dead, I can't any more.

Just wondering why this was moved to bugs? I'm pretty sure it's just an Easter Egg. I've know about it for years.

Just wondering if many people knew that if, after ejecting and landing, you press the eject sequence again, your pilot drops dead?

Hey there. I would be delighted to see a version that will run on Android. I understand developing software no trivial act and given that I'm unlikely to change platform, I would be happy to pay. Oh... and of course Black Shark 2 while you're at it. And then do my taxes.

Hi Black Sheep. I see two issues of contention in your setup. First, an older video card with only 512MB VRAM is definitely a bottleneck. Having used a RADEON HD4850/512MB myself with DCS A-10C I had no option but to reduce texture detail Medium to alleviate the graphical bottleneck. With the decoupling of world and cockpit texture resolution in the latest 1.1.0.8 patch, Low texture detail may even be an acceptable compromise. I would also highly recommend reducing water detail to medium and disabling HDR. You should get a decent experience at your specified resolution without crippling quality compromises. Furthermore your use of XP Pro is also a limitation, assuming it is the standard 32-bit version. Check Mustang's signature for the forum threads detailing workarounds for the OS's memory limitations. Upgrading to a 64-bit OS will significantly improve the game's performance and stability (going from WinXP/32 to Win7/64 did wonders for me), although I believe the latest 1.1.0.8 patch goes some way to address this issue as well. Good luck.

I believe the TRIM command will be utilised regardless of bus mode. The main factor in TRIM support is the OS, which I think has only been natively implemented in Windows 7 to date. The most direct benefits of AHCI are hot-swapping and NCQ support, the former being of little use to an internal drive and the latter being of little use to a SSD. From what I've read in the Microsoft forums, disabling the swap file is stongly discouraged. While the OS will work, it has been designed around the assumption that the swapfile exists and will be utilised. Even with a Terrabyte of RAM, Windows may still use the swapfile to a limited extent, although I do agree that the best way of minimising this is to install as much RAM as is reasonable. Furthermore, I agree with topdog and would argue that the SSD is by far the best place for the swap file and the argument that it consumes too many write cycles is highly subjective at best. See this post for a more detailed explanation. I disagree with this entire statement. Small, random IO is what a SSD excels at and so, like the pagefile is an ideal destination for indexing. I have never encountered the argument that indexing degrades a SSD's performance before, I'm not aware that it is even technically possible. Indexing is designed to run in the background only under idle conditions. As for exhausting write cycles, I do not believe that the amount of disk writing that indexing performs is sufficient to reduce the drive's write-life by any considerable factor. I refer again to the link above. As for disabling indexing, all I can say is that it's solely a matter of personal preference.

In terms of performance, with an Agility 3 or similar SSD there would be very little benefit in multiple physical drives unless you wanted to get fancy and put them in a RAID 0 configuration. The random IO performance is so many orders of magnitude higher than a mechanical HDD that it will handle multitasking with ease. I use an original Vertex (Indilinx controller) for my OS, pagefile and A10 with no issues at all. With the exception of the very first generation, SSDs and their OS support are mature and reliable enough to use without many special considerations.

I'd love to see a kneeboard like a few other sims have that the mission designer, pilot or simulation can fill with relevant information such as charts, pictures, mission objectives, frequencies, landing airstrip and ALA details, important comms, 9-lines, etc, etc. This would facilitate the collection and organisation of important mission information without requiring the pilot to go do so by hand, thus reducing the amount of pilot load and allowing the pilot to get into the mission faster. The option to disable this could be made available to pilots preferring the manual method.

I was going to post on this very issue myself. I agree with EbonySeraphim that mission progress/completion feedback would be appreciated. It would be good to get at least a "Mission Complete" after one lands and exits the simulation. At present there is absolutely no compelling reason for a player to actually land the plane after completing the mission objectives beyond that which we impose upon ourselves.

I'm generally an ATI/AMD user myself (although I have used the occasional NVIDIA card) and I've never had any issues with the ATI Radeons or their driver suite. As you mentioned, both cards cost and perform roughly on par with one another so as already mentioned the choice comes down to personal taste as each will fulfil the role adequately. I believe however, that it is the Radeon 6950 that consumes less power. Anandtech

I just want to say that, as somebody who has researched a bit on SSDs out of both personal and professional interest as well as being an SSD user, I would like to state that the while the notion of moving the pagefile off the SSD to prolong its life is often mentioned around forums, that I believe it to be a highly subjective truth. I have a 60GB OCZ Vertex (original) upon which I keep my OS, documents, pagefile, a virtual machine and a few high-performance games. As a software developer I have made thorough use of the drive (by typical desktop standards) and after over a year of use, my drive lifespan sits at 95%. This works out to approximately 20 years of drive life (based upon write-cycles only) which outlives anything we generally use a hard drive for. So while technically, you will improve your drive's write life by moving the swapfile elsewhere, when considering the timeframes involved with typical desktop use, not only do I believe this to be unnecessary, but one also loses the performance advantage of having their pagefile on an SSD. The most cost-effective way to reduce pagefile dependence is to increase your system RAM. My strong reccommendation is to move your OS to the SSD. You will find the loading times and general performance improvements to be astonishing. From experience I highly recommend you make the upgrade if you are able to afford it. Barring some new technological breakthrough, I would never go back to a mechanical drive. Out of curiosity, have you considered a Sandforce based drive such as the OCZ Vertex 2 or 3? The Vertex 3 significantly outdoes the Intel drive in terms of both performance and (at least historically) price and is very well received in professional reviews.

Oh yeah... and a Vertex 3? Lucky bastard :)

Well, as a SSD user, as far as I'm aware there isn't too much that you can do to completely fizzle a SSD just through using it. If you drop it in water or use it as a lightning anchor, well that's another matter. This is news to me. As far as I recall, Windows XP has potential issues with partition misalignment causing limited performance loss however in general a full format won't hurt the drive, aside from wasting a write cycle. If you mind my asking, where did you hear this? Yes, no need for a full format, the drive handles internal errors, bad blocks, etc quite well on its own. No different to a mechanical drive. Defragmenting is of no value to a SSD. It accesses data in a very different manner to a mechanical drive so gains very little benefit from a defragmented partition or file relocation. The only thing defragmenting a SSD will achieve is to needlessly reduce available write-cycles. Exactly the same as a mechanical drive. Really you only need to blow away the existing partitions. Only do a full format if you absolutely want the security of removing remnants of old data which, if your SSD and OS support the TRIM command (yes in your case) may be a moot point anyway. Hope it helps.

A query with respect to online squadrons: Do you get more happy-go-lucky sorts that accept and train casual players or do they tend to be more interested in the serious simmers and selective about their membership and time investments?

That was my same experience when I played around with Falcon 4 AF, although there was a bit more tactical info in the manual, it basically left responsibility for learning the practical applications of the aircraft up to the user. My question is why this assumption is made? My guess is the large amount of effort required to compile a suitable set of documents and training lessons, but I can't believe it to be so complicated that at least some basic documentation can't be included.

Hello folks, While you'd think a person would be satisfied with a 650 page manual, I have to admit that I'm really disappointed by the complete lack of some critically important content. Where are the instructions (or in-game tutorials for that matter) on actual combat in the A-10C? I've gone through the bulk of the manual and have been enjoying myself learning how to 'fly' the A-10C and deploy weapons but now many important questions are circling my mind. How do I learn to fight? To manoeuvre? What is a barrel roll? How do I perform one and when do I use it? How do I avoid SAMs and evade launches? What weapons do I use to attack a moving column of tanks? What altitude do I attack from? How do I coordinate with other aircraft? I realise that compiling such a set of documentation is not trivial but there is an enormous component of the simulation that a novice to military simulation as myself is left completely without direction.