Rhinox

Of course there is one method (as easy as assigning keybard button): simply plug in your THW (I recommend both joy and throttle), go to TIR profile editing, click "assign new button" (or something similar, I do not have it here) for "pause" and click desired joy-button. And instead of "F10" (or whatever it is in default settings) you will see something like "Joy8". TIR-software captures even joystick-keypresses, so there is no need first to assign some keyboard-key to joy-key. I did it this way some time ago when I was using x52, but it should work with any game-controller recognised as directx-peripherial (now I have THW and using TARGET, which imho is worth to learn!)...

Honestly, I do not thrust these remote-control solutions like LogMeIn or TeamViewer. What I do not like is that traffic is routed (or re-routed) through software-company's servers. So your connection is not "PC->YourServer" but rather "PC->LogMeInServer/TeamVieverServer->YourServer". You can imagine yourself what security implications such a configuration has... I prefer direct solution, like VNC (UltraVNC, TightVNC, etc). It gives you the same possibilities of remote-control without any intermediary.

Right now (18 december 2011, 17:57 UTC) there is something wrong wich checksum of the first file of DCS:BlackShark 2 1.1.1.1 (english download version). Website says: File Name: KA-50_full_en_1.1.1.1.exe File Size: 1.85 Mb MD5: a6c8c0dca2e1442f38faffa3c094dfc0 But this file has MD5-checksum: 732e8150158415ab8bb5b92c3e0f0ce1 I suppose this is simply copy&paste error on the web-side (because there is the same value of md5-checksum for both KA-50_full_en_1.1.1.1.exe and KA-50_full_en_1.1.1.1-1.bin), but I recommend to postpone downloading until this issue is solved.

True. For me (i5-2500k@4.6GHz, HD6950->6970) cpu is bottleneck in LO:FC, but gpu in DCS:A10C. But it also depends on settings, i.e. increasing AA/AF puts higher demand on GPU, but "civilian traffic" takes som extra CPU-time...

Single LCD 1920x1080, default AA/AF (application controlled)...

It depends on your settings. My HD6970 runs 100% all the time while playing A10C. I tried to overclock my cpu a little higher but I did not get a single fps more. On the other side, in LO:FC it is never loaded higher than on 50%, even with every possible setting on max/full...

AFAIK there are at least 3 versions of them, so pay attention to which "GTX560Ti Twin Frozr II OC" you pick: A: 1GB/GDDR5, 384 stream-units, 64 texture-units B: 2GB/GDDR5, 384 stream-units, 64 texture-units C: 1.25GB/GDDR5, 448 stream-units, 56 texture-units

For gaming-notebook (if there is something like that at all) GTX570/560M or HD6950/6970M is imho reasonable choice. Nothing less makes sense, especially if you want to play graphics-intensive games. You have to realize mobile-gpu is not the same as equally numbered desktop-gpu. For example GTX560 (desktop) is ~two times more powerful than GTX560M (notebook). Or in other words, from performance point of view GTS450 (desktop) is as fast as GTX560M (notebook)...

And what did you expect from this??? Man, GT550M is at best only "lower-middle" class in mobile-GPU. When we take 3dmark06/Vantage score into account, it is about as "powerfull" as desktop low-end GF240GT. Absolutelly unsuitable for mobile gaming...

Maybe. It also depends on overall airflow in your case. But I'd recommend to invest at least in better-than-stock air-cooler if you want to touch 4.5GHz (i.e. yasya/mugen for ~30-40 bucks, I'm running i5-2500k@4.6GHz with it).

I came accross a few rumors concerning AMD's future plans, and finally found the original article: http://www.mercurynews.com/business/ci_19358655 Quote: "We're at an inflection point," said company spokesman Mike Silverman. "We will all need to let go of the old 'AMD versus Intel' mind-set, because it won't be about that anymore." Some comments speculate this might mean shifting of primary business from pc/cpu (and maybe gpu) to somewhere else, probably mobile-cpu, tablets, smartphones, etc. Honestly, the latest AMD "bulldozer-architecture" seems to be nothing better than failure, but if AMD leaves pc/cpu market, that will be very bad thing for all of us (being "AMD-positive" or not). Monopoly never meant anything good to customers; competition did. So let's hope these are just rumours...

Download & install "AMD system monitor", start it, then DCS:A10C, play a little, leave DCS and check diagrams. The line close to 100% load is your bottleneck (be it cpu, gpu, or ram). I think it is your graphics-card...

IIRC, gameport has analog interface where each axis value is actually resistance 0-100[kOhm]. So you can not wire two in series. It would be 0-200[kOhm], so your pedals would work only on half way. But I do not see where the problem is. Use simply gameport/usb converter, and you get 2 more axis (in addition to your stick), assign them for brakes. LOFC does not use axis for wheel-brakes, so you'd have to program some macro for it. And I think A10C needs two axis...

CPU: Try to overclock cpu not "just a bit" but as high as you safely can (my 2500k is running at 4.6GHz and yet it not too fast). Use this thread (Z68 is similar to P67) http://hardforum.com/showthread.php?t=1578110 GPU: I'd recommend 2GB-version. It costs about the same, and 1GB might be not enough for DCS. And I think it is worth to buy HD6950 with "standard" pcb-layout, not those IceQ/Frozr/Flex versions. You can easily unlock it to full HD6970 by flashing new bios (which is easier to find for standard HD6970/2GB). It is pretty simple and safe, because HD69xx have actually dual-bios. You can have 6950 in bios1, and 6970 in bios2...

What drives me mad is that auto-switching from radar to IRST on MiG29/Su27 when target-lock is lost. This is not bug, but rather ULTRA-MEGA-BUG! Not long ago I visited one airbase and have been talking with MiG29 pilots, they confirmed IRL it never gets auto-switched from radar to infrared... But first of all, I'm giving my vote for track-bug. Inconsistency between what actually happened in the flight, and what track is showing during replay is something I can not understand at all. Some server-admins request track in case of complaints, but what the hell can such a track prove?

I have tested DCS:A-10C on HP/dv7 (~900€, i7-2630, 8GB RAM, HD6770M) and compared it with my PC (i5-2500k, 8GB RAM, HD6950). Well, tried to compare, but honestly it is uncomparable. With the same settings laptop gave me ~30-50% fps compared with my desktop (without overclocking, moreover laptop was running at 1600x900, desktop 1920x1080). Although components look similar, in reality it is completely different class. Those "mobile" cpu/gpu are waaay slower than desktop equivalents. I doubt there is anything like "gaming laptop" at all, maybe some "Alienwares" are close. But still, for the same price you can get desktop-rig twice that powerfull... I do not say you could not play DCS:A10C on laptop, but you'd definitely have to meet some compromises, and be very conservative with settings. You can get "playable" fps if you turn settings to min/low/off, but it is up to you whether you are satisfied with it. Call me mardy, but I was not!

I observed this issue too, but in my case it is a little different: sometimes it does not happen for a few days and our server is always in the list. Other times I do not see our server in the list right from the beginning, and even restarting the server does not help. And what is even more interested, it is not consistent for all users: sometimes some of our squad-members see the server, others do not. It is a kind of random behaviour, changing from time to time. But at any time we all can connect using IP (fyi, you can even connect using domain-name, it is easier to remember).

I do not agree with your last statement... ;-)

OMG Eth, do not bull***t us please. Upgrades like this will happen in the future too, and you know it as well as me. If flightsim wants to survive and remain attractive for users, it must be updated at least every two-three years. Technology is advancing, what we see today as state-of-art becames obsolete very soon. And what will make ED to stop charging customers for such upgrades in the future, if they do it now? They will need sufficient funds to work on those upgrades, especially when number of necessary compatibility-patches increases with every new DSC-release... I'm pretty sure that's exactly what most of us would see as welcomed and acceptable strategy: paying let's say once per year or two for core-upgrade, and for every new DCS-module when it is released. But *not* charging customers for every compatibility-patch! Because once in the far future, when DCS consists of 20 releases, users will have to pay $50 for new DCS:No21, and 20x$20=$400 for 20 "upgrades" of DCS:No1-DCS:No20 (not to talk about time necessary to create/release those 20 upgrades)...

I would not have any problem to pay ED for their hard work, had I not the feeling they artificially made up some job just for earning more money from us, customers. Let me put it straight: *IF* DCS followed modular principle (one common DCS:core and pluggable DCS:addons), I would *gladly* pay once full price for DCS:core, later for DCS:core upgrade every time it gets upgraded, and for every new DCS:addon (but only once per DCS:addon, and fixing eventuall errors/problems by addon-patch schould be cost-free). But DCS is more like a series of independent software packages, and thus amount of necessary "compatibility/upgrade/patch-development" is increasing exponentially with each new DCS:part. I understand this is difficult task but I think this work on "each-with-each" compatibility/upgrade could have (and should have) been avoided. Otherwise with each new DCS:part we will have to pay more and more for upgrades/patches of all those previous DCS:parts...

Could not put it better myself. I do not care of a few bucks more or less (just bought full BS2, instead of just BS1->BS2 upgrade). But I do not want to be handled as a cash-cow. I think ED-customers deserve better treatment... And concerning this move from modular DCS-system to more-or-less independent packages, I'm not sure it is a step in the right direction. Instead of having "only" two tasks (1. developing new DCS-module, 2. updating DCS-common graphics/physics/maps/simulation-core), in case of independent DCS-packages ED will have to update/upgrade each of them, when new DCS:part is released. It is maybe not so obvious now when we just have two parts of DCS-family, but will became clear in future. Imagine we have 20 DCS-planes, now ED just releaset DCS:No21, and we have to wait&pay for 20 compatibility-patches/upgrades (instead of just buying/paying for DCS:No21 and one upgrade for DCS-common core). And it is not only about money, but also time is a factor. With more and more DCS-parts we will have to wait longer and longer, till they are all finally online-compatible with the last new one...

Correct me if I'm wrong, but I remember DCS:BS being presented from the beginning as a modular system, where each new module will be compatible with those previous. Do you find it correct, when customers now have to pay once more to get what was promised to them?

Thank you for answer. I just wonder why is this very important info not mentioned in announcement? (Answer: because users would hardly agree this being "improvement" over BS1)

So DCS:BS is online-compatible with DCS:A10C, that brings me to this question: How is it with DCS:BS and "master server"? Does DCS:BS2 need it too? Or only DCS:A10C? If DCS:BS2 does not need "master server connection", can it still host mixed missions for both BS and A10C? (sorry if this question has been already answered, I tried searching the forum but found only a lot of thrustmaster crap in signatures instead)

I tested Supermicro 815/1U when building my server, but I would not recommend it for FC/DCS or any other application where dedicated gfx-card is required. Apart from being expensive as hell it was nearly impossible to guarantee sufficient cooling even for very modest gfx-card (as it is for every 19"/1U). If you are not limited with space, use at least 2U-case instead. More space for drives/fans, better cooling, and you can use virtually any stock standard-size components (even dual-slot monsters like HD6990/GTX590 with riser-card, though I think it is overkill, I'm using low-profile HD6570). Many colo-centers charge you more for bigger cases, so I think 2U is quite a good compromise...