oyaji

Free Play Account
  • Content count

    8,194
  • Joined

  • Last visited

  • Days Won

    1

oyaji last won the day on August 24 2016

oyaji had the most liked content!

Community Reputation

0 Green Tag

About oyaji

  • Rank
    Member
  • Birthday

Profile Information

  • Preferred Side
    Allied
  • Preferred Branch
    Navy
  • Preferred Unit
    River Boat
  1. nobody really quotes pixel fill rate anymore. take nvidia's top card as an example 49.4 billion textures per second to get the pixel fillrate you use the following coremhz * rops = pixel fill rate. the gtx580 has 48 rops and has a mhz of 772 so 772 * 48 = 37.05 billion pixels per second. even that isn't that important though as not much is constrained by raw fillrate these days. in ww2ol the only thing GPU limited appears to be fps when zoomed into trees which is heavily shader limited. apart from the trees ww2ol is still heavily reliant on the cpu. you can test you cpu performance by typing in .benchremagen in the offline client prior to spawning in. to test your shader performance i suggest going offline and using the forest that is NE of the rifle firing range. the nvidia 260 and ati 4xxx line of cards are really all this game 'requires' for good zoomed in to trees performance. ~3ghz AMD/core2 or ~2.6ghz corei5/7 gives you good enough fps everywhere else. more obviously is better. yeah it's all about the shaders these days, not like the old days I still favour nVidia myself, but there are pro's and cons to any choice I'll try and get a bit of a run down presented as an article on our front page after beta is all done with But WWIIOL does not take advantage of multiple GPUs, does it? But will WWIIOL do so in the future? And if so, what is the approximate timeline for that? Based on texture fillrate numbers, ATI trounces nVidia currently. ATI's best single GPU texture fillrate is their HD 6970 (MSRP $369) at 84.5 GT/s while nVidia's best is the GTX 560Ti (MSRP $250) at 52.6 GT/s. The price difference there is quite large, but compare the HD 6950 (MSRP $259) with a whopping 70.4 GT/s fillrate. The texture fillrate advantage of ATI is currently enormous. but ATI has a horrendous history of piss poor driver support, and nVidia noticably better like I said, all choices have pros and cons Wanna hear sumpin' funny? The 9800GT texture fill rate is about 34 GT/sec and the 460 is around 37 GT/sec - nearly identical! And the old 9800GTX+ pumps out 47GT/sec - more than the best of the 400-series! Look here: http://en.wikipedia.org/wiki/Comparison_of_Nvidia_graphics_processing_units One fella told me that the DDR5 RAM makes a big difference over the older DDR3 on board the 9800GT. But performance numbers are listed as nearly the same between my card and the 460. So which do I believe? and this is where you realize fillrate means almost nothing these days. its all shader count and shader speed. gpus have been able to fill 60 frames at 1920x1200 for many years now but that doesnt mean anything if you can't run the shader operations at only 20fps. fill rate = dinosaur performance metric. ddr5 does make a fairly significant difference as it is quad data modules. two reads/writes at the rising and falling edge of the clock cycle. hopefully similar memory will find its way to the system bus soon. usually GPUs have been a step or two ahead on all the major ram breakthroughs. If you read back through the whole discussion posted above your reply, you'll see that the whole discussion has been zeroed in on one specific, texture fillrate, and that it seems to imply that the performance of cards with regard to "shaders" (whatever that means) is determined by texture fillrate. Is your reply on target or is it a non-sequitor? Is texture fillrate the metric that determines vid-card performance for WWIIOL or not? ...
  2. Sounds familiar to me too... I'm pretty sure I was trying to record at 30FPS and couldn't do any better than 11. I am also pretty sure I went for either low or no compression. That costs more in storage space for less processor overhead, unless I am mistaken. (Besides not wanting to waste processor cycles in compressing video while recording in real time, I would always prefer to have raw video at full resolution and compress it later if I need to - as a matter of quality). My connection isn't bad these days - it improved after the local telephone company upgraded their equipment after all the hurricanes we've had over the past 5 or so years: I usually ping around 50-60 with 5MB/sec download and 1MB/sec upload capability to Dallas according to SpeedTest (although I have seen ping time double and data transfer numbers drop to half on occasion). Packet loss is usually a non-issue. I still think that if I could get my combat framerate up higher that I'd then be able to record at faster framerates. Doesn't framerate always drop when you do HDD recording?
  3. I don't have to look it up - those are Western Digital 500GB drives each with a 32MB cache. As I mentioned, they are in a RAID0 on 3.0GB/sec SataII, and the only thing they are for is recorded video (meaning they haven't ever been used for anything except 2 trial of video recording). I thought that would be a pretty good setup, but I ended up disappointed. I used a 2-week free trial version of Playclaw for my recording attempts. It was about 8-10 months ago that I last tried recording and then gave up in frustration. I don't remember for certain if I tried setting the affinities of WWIIOL and Fraps to 2 different pairs of the 4 cores in my processor, but I am pretty sure I did - assigning affinities has been a habit of mine for about 5 years so that I could always try to squeeze best performance out of the game using bottom-end equipment (until last year I was still using an 8xAGP nVidia 6800GT card - and only went to PCIE when AGP became out of date for the game). The AMD 925 is not Black Edition. The motherboard is fairly basic too, but does allow some sort of rudimentary overclocking. I figured I might have to learn a bit about how to get some extra boost out of that when I have run out of other options and when I have a better heatsink than stock (I was thinking one of those Zalman "hoop-style" jobbies would be a good way to go... they look like they might do a better job mounted horizontally than the typical rectangular heat-pipe layouts.) How much did overclocking your 955 from 3.2GHz stock up to 3.7 do for you? I am wondering if the vid card memory at 512MB is the problem, or if it could be the Northbridge (it does not have heatpipe cooling, only a rudimentary finned aluminum heatsink). Your 5770 has 1GB of video RAM - does the 8800 also have 1GB or is it 768MB or 512MB? ... One last thing - when I stepped up to a nVidia 9800GT from a Radeon 3870, from a 6800GT (AGP card), I noticed my air-to-air kills jumped up each time along with my framerate. I've had a sneaking suspicion that automatic weapons fire hits go up with increased framerate - and that sorta confirms my suspicion that having a crappy framerate just allows the enemy to fly through my bullets that do not render in-between frames. Seems to me that K/D and fighter prowess can be bought to a fair degree in this game, which goes against the grain with me since I started playing this game 10 years ago because the idea of a level playing field: that standings here are a measure of player ability and not the result of artificial measures that result from "levelling" so common in other games. I think I do fairly well with the equipment I have, but vanity compells me to want to do better. I am neither willing nor able to spend much on that endeavor, though. Does it really make much difference?
  4. What resolution? Not sure why Playclaw would be killing it that bad considering your system specs. Either there is a driver issue or there's a hardware bottleneck at the vidcard. I've a budget system: i5 750 intel 2gig ddr3 1333 Biostar T5xe mobo ATI 4860 single 300gig sata drive 1680x1050 with most settings on high I can still run playclaw and hold over 30 fps around contested towns. I am on 1680x1050 resolution, 32-bit color. Everything else is turned off or turned down to minimum settings (except tracer smoke and muzzle flash). By the way, the 9 to 11 FPS framerate was over open sea at 3km and only 3 planes in sight. That is freaking unplayable. ... Unfortunately, the series number, especially the first digit in the 4-number series designations, is less and less meaningful. I recommend always checking Toms Hardware. Looks for the most recent "Best Graphics Cards For the Money" article (they do one each month). http://www.tomshardware.com/reviews/best-gaming-graphics-card-geforce-gtx-590-radeon-hd-6990,2879.html Check the last page of those articles, which is the Graphics Card Hierarchy Chart http://www.tomshardware.com/reviews/best-gaming-graphics-card-geforce-gtx-590-radeon-hd-6990,2879-7.html BLOO - note that I wasn't making much over the model numbers, but instead just named them to contrast the texture fillrates that I looked up on all of the on the Wiki comparison charts (link provided in original post quoted above). If "[texture] fillrate is king" for WWIIOL, then Tom's Hardware comparisons are too general. Zeroing in on the texture fillrate numbers should provide the best comparison, right? So does the type of graphics RAM - GDDR3 versus GDDR5 - make any additional difference? And then there is the amount of RAM - typically a comparison between 512MB, 768MB, and 1GB. How much difference does this make? Would it be worth grabbing a card (that is used only for this game) that has 1.5, 2GB, or more GRAM so that the card performs better now? And how about the future - with an eye for coming changes to the game? When I pony up for a graphics card, it had better last me at least 3 years. I'd love to make some instructional (particularly bombing) videos for the game, but I cannot afford the gear for that. This current machine is the closest I've ever come to it I think - how far from it am I now? Last time I made an effort was 6 years ago when I spent $1,650 getting a machine that would be capable, but it wasn't. In fact it was only competitive for air combat without recording video about 6 months - after that I was just farther and farther behind every new patch. If it wasn't for wealthier players tossing me their hand-me-down gear I'd have been forced to quit a couple times in the past. How much do I need to budget per year to keep playing this game at, say, the 80th percentile level of the air war? ...
  5. Wanna hear sumpin' funny? The 9800GT texture fill rate is about 34 GT/sec and the 460 is around 37 GT/sec - nearly identical! And the old 9800GTX+ pumps out 47GT/sec - more than the best of the 400-series! Look here: http://en.wikipedia.org/wiki/Comparison_of_Nvidia_graphics_processing_units One fella told me that the DDR5 RAM makes a big difference over the older DDR3 on board the 9800GT. But performance numbers are listed as nearly the same between my card and the 460. So which do I believe? ...
  6. That is a really good point. I'd like to get a better framerate in the game but in air combat I am getting 40-60FPS (on minimum settings) For me the burning question is "what do I need to get good enough performance to record video with PlayClaw or Fraps?" I've been wanting to do so for years (in particular for instructional game vids), yet with this computer I've fallen flat: AMD 925 Phenom II x4 quad-core processor (2.8GHz) 4GB DDR3 1333 RAM nVidia 9800 GT (512Mb) graphics card M4A785TD-M EVO motherboard 300GB PATA HDD for system and applications plus 1TB RAID0 (2x500Gb drives) on SATAII (3Gb/sec) interface I get just 9 to 11 FPS recording video. Where is the weak link - the vid card? Would I be better off just buying a cheap DDR5 card, or do I need to wait and buy a HD 6950? Whatever card I get needs to last me a couple years at least... ...
  7. ... But WWIIOL does not take advantage of multiple GPUs, does it? But will WWIIOL do so in the future? And if so, what is the approximate timeline for that? Based on texture fillrate numbers, ATI trounces nVidia currently. ATI's best single GPU texture fillrate is their HD 6970 (MSRP $369) at 84.5 GT/s while nVidia's best is the GTX 560Ti (MSRP $250) at 52.6 GT/s. The price difference there is quite large, but compare the HD 6950 (MSRP $259) with a whopping 70.4 GT/s fillrate. The texture fillrate advantage of ATI is currently enormous. ...
  8. A friend has returned to the game and is having sound problems. The game sound is fine until he starts Teamspeak2, whereupon TS2 works just fine for him (from the front panel jacks, if I have the details right) but he loses all game sound! Any suggestions for him please?
  9. No complaints about the rules, everyone who signed up to be a target on the airfiled knew what he was getting into. On the other hand, spending and hour and 30 minutes waiting to get shot on the airfield (or in my case, as one of the "lucky ones" who made it off the ground, getting shot 3 minutes later before ever seeing the attack) turned out to be not so much fun. I won't be back for next year unless there is a better chance of meaningful action - playing the sitting duck, hoping to survive straffing/bombing, hoping to survive the climb out and fight up to enemy fighters in inferior climbing machines, and hoping to get a crack at the bombers after all that just isn't worth it: the odds are crap for so much time spent waiting. None of our 9 fighters (of the 12 who started out and actually waited the whole hour and a half) had a chance to attack the swarm over the fleet. Our 2 fighters who did actually make it there were so crippled that they were unable to get the altitude they needed to attack against the 30 to 1 odds versus bombers and 2.5 to 1 vs. fighters even on the second wave, and spent their time dodging bullets and shooting none. If you cut the number of fighters and then placed a hundred clones on the airfield, maybe that would improve the odds enough to generate future interest. As it was, the Allied planes were just part of the scenery and there to provide themselves as targets for other people's fun.
  10. There were only 9 Allied airplanes instead of 12 (some guys got tired of waiting for an hour and a half and went to play on the live server instead), and of those only 4 I know of survived to get off the ground. The H75 was Ratware. When I was foiled by too much speed to land rounds on you, I bled off too much, trying to get a quick save for Ratware (whom you damaged on the first pass). I should have played it safe and extended out level, but I wanted to save my friend and so I got too aggressive. One BB plink headshot and it was over - an hour and a half sitting waiting for the event spent for 3 minutes of getting shot on the airfield, struggling to climb, and 2 bursts fired at bad angles - bah, not the best time I've had in the game, to be sure. The FW190 who straffed and killed my friends Adam and Dasquirrel in their fairmille looked like he was having fun though. Ratware lasted through the second wave with his damaged plane ony able to make 260kph and unable to climb, so the 109s and the 190 had an easy time with him - crippled, he spent all his time dodging bullets and not getting to shoot at anything. Since he wasn't allowed to change his plane to a H81 even though we had no-shows and were down 3, nor could he replane either by the rules, he just stayed to offer himself as a target. I don't think that was very much fun for him, either. The reason you couldn't find any EA was because there weren't any. I was gonna split, but I was reminded that I was allowed to play AAA. Being a pilot I had not the least interest in that, but I did wonder what the battle I missed looked like, so I wheeled out the fastest AAA gun, a Ca mle 38 and pushed out to the water in time to see the second wave. I tried to shoot the planes off Ratware but missed the FW190 that was after him. I did get one of the 109s that was on him - about 2 seconds after he tore off Rat's wing. On our side, we did have a P-38 who kept spawning in and who finally left and never did take off (after I asked him several times not to, because the event was going on) - and one side-switcher who spawned an H87 a couple times who did. I hope he didn't kill anybody, but I don't know. I also bagged a Stuka, so I guess it was worth pushing out for the second wave even though I don't know how to do anything but fly. I was cussing that damn gun the whole time - dead slow and no climb rate at all, I won't take one of those out again! I never got to see the battle, except for about 20 seconds of about 70 Stukas appearing straight up, diving straight down, dumping all bombs on 1 pass, and skeedaddling. It sure was a lot of "hurry up and wait" for very little action. It must have been more fun for the ships and the bombers.
  11. That's good to hear, especially that you want to organise. The Ad'A planes are less than half as popular as RAF, so there aren't that many players flying French: the Allies can sure use the pilots. Have you decided who will be your Squad Liason Officers? I suggest at least 1 for every timezone during which you will fly. Best of luck to you - if there is anything I can do to assist you, let me know.
  12. H81 for me, might also fly a H75 - or nothing, thanks.
  13. Long history in the game, and some very fine pilots have flown with the Eagles Squadrons, some I've flown with were Manhi, Rappy, Dair, Kifaru, Sara... and of course my buddy Rivait. I hope to have the chance to fly with them all again. And so too should you. If you are looking for a good outfit to fly with, the Eagles are among the best.
  14. It IS locked. I am a trainer and I cannot get in either.
  15. How about a maybe for something RAF, if I am free.