Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by HappyHaddock

  1. "Curiouser and curioser said Alice" Well... Having tried my old RAM in every configuration of the four slots in the mother board it would only recognise 8GB of the total of 16GB whatever I did so I assumed I'd had a hardware failure on one of the paired sticks of RAM. Consequently I've ordered another 16GB (2x 8GB) and now they have arrived in the post I've taken out the old and put in the new. It took a little while to boot up the first time as presumably various registry files and other such memory related stuff were updated by Windows, but by the time it had opened my desk top it was running smoothly and quickly again and actually showed as having 16GB of RAM... All good and well. I know there is various debate about the benefit that running memory in paired dual channel has compared to a single stick of the combined total capacity of two matched sticks, and that where the mother board supports it (which almost all do) it is generally felt better to go with paired. However the general consensus is that if you have a spare stick of RAM lying about it does no harm to plug it into a vacant slot even if it can't be accessed as quickly as had it been paired as dual channel with double the band width... So rather than waste the one remaining good stick of old RAM I thought I'd install it and take my total up to 24GB. Well damn me if I can work out which of the old sticks has failed. The two old sticks installed together without the new RAM are detected as 8GB, either one installed on its own is still showing as 8GB and all four sticks (both new and old installed together ) show up and run as a total of 32GB. God knows what is going on but I now have a PC with a gratuitously excessive 32GB of RAM way out of proportion for the needs of a 7yr old PC. Anybody have any suggestions as to which strings the gremlins are pulling the background? HH
  2. I can't answer for LizLemon, but when I've needed to edit and change mgm files I use a basic hex editor like HxD or HexEditorNeo. Basically they just display the raw code, a lot of which appears as unfathomable strings of seemingly random characters but with patience trial and error it's surprising how much you can start to make sense of in order to change things. HH
  3. Thanks for the reply, Windows own memory check utility mentioned in your first link is something I had already ran three times when trying to get it to recognise the second stick of RAM but it would not recognise it, it just kept saying there were no faults with the 8Gb it could detect. As for other utilities I've tried various that supposedly track RAM usage and speed, but so far none have found the other 8GB to actually monitor it. So until the new memory arrives I'm running half the memory I used to have, and running in single channel format may further slow things down compared to when it was running as paired dual channel. It's been quite noticeable that various other bits of "more demanding" software have been slow to open over the last couple of days, and most particularly they have been slow to populate lists when clicking on any kind of "file open" tab or "save as" tab, and even something "simple" like selecting a font/typeface in Word is taking a little longer to populate a list of choices from the moment you click the mouse. Any kind of slow down on retrieving data does hint at memory problems so I'm hoping that one stick of failed RAM is the only source of my problems and it's not indicative of bigger issues? HH
  4. Well I could have done without a day chasing diagnostic work but... After wasting a lot of time addressing a corrupted windows update that failed to apply (second one in almost as many months) and getting that sorted the problem still remained and so I spent time focussing on various memory checks. Not a single ap or memory check utility reported any fault, every one reported 8GB of RAM working faultlessly... however my machine has 16GB installed, except windows and Bios can only find 8GB to test. For a PC the age of mine 8GB is probably all that most would expect it to have and it seems fine for running most regular activities, but it seems likely there has been some sort of hardware failure on one of my 8GB sticks of Corsair DDR3. Quite why (or even if?) this is the cause of the strange way that the FPS would drop with subsequent re-loads of the same mission in IL2 I don't know, but for the relatively modest costs involved I'm going to order and replace all the RAM in my PC and use it as a chance to upgrade to the fastest DDR3 the motherboard will take. Then I'll start again with new tests to see how my PC behaves and if replacing the RAM has cured the issue? In the mean time is anybody aware of a utility that can actually detect faulty/failed RAM to conclusively prove it is a hardware fault with the RAM, and not some setting somewhere that has "disabled" part of it. I've switched the chips around in the various slots on the motherboard and which ever of the four slots I plug the two 8GB sticks into it sill only detects a maximum of 8GB so I'm fairly confident it s a hardware failure, but it would be peeving to replace the RAM and find my PC will still only detect 8GB despite having a maximum capacity on the motherboard for 4x8GB, and when until recently it ran fine with 16GB. HH
  5. Well I've tried a few more tests and it is looking like the most probable culprit is Graphics card memory but that is far from conclusive. There was a mention in some of the threads linked above that Vsync might help and it sorts of makes a difference... I initially thought having Vsync turned on was solving the problem as after five re-launches of the same mission, aside form the first second or so of each re-launch having a wildly fluctuating FPS it settled down to where I'd expect it based upon how the mission runs the first time I load it. However by the sixth re-launch it failed to settle at where it should be and remained at approx. half the expected FPS, subsequent re-launches of the same mission continued to show the problem. I tried the same expt. on a much earlier and less technically demanding map. I don't know if having Vsync on actually solved the problem there or simply "bought" me enough extra re-launches of the mission that I got fed up waiting for the next re-launch to show the drop in FPS... On a simple mission just sitting on an empty runway on the Lapino map the sim started up multiple times, each one at around 70+FPS with Vsync turned on, whereas without V-sync by about the third or fourth launch I was seeing that drop to about 40FPS. The same test of starting up a mission just sitting on an empty runway on the Kuban map would first launch at 45+ FPS. With Vsync on I got about five or six relaunches before FPS dropped to about 20. With Vsync turned off I only got about 3 re-launches before I saw the same FPS drop. HH
  6. Thanks folks for chiming in with various thoughts and ideas, sometimes reports of technical probs don't get the same response from the community here as posts with pretty pictures and exciting videos... Anyway it is both reassuring and alarming in equal measure to hear that this is an issue affecting others, reassuring to know it's not that there's something inherently wrong with my PC if the same problem can affect new PC's, but alarming that there's no clear consensus about what needs to be done to solve the problem. HH
  7. Well from the replies here it's sounding more like an issue particular to my PC, or at least older PC's if others with newer computers can fire up the same mission time after time without experiencing any loss off FPS. My PC is now around five years old, possibly a touch more (i5 3570k 16Gb ddr3 RAM gtx970). I tend to favour running all graphics settings maxed out and when I run aps that monitor performance during IL2 the four cores of the CPU tend to run at 30-40% usage rarely spiking above 50%, memory tends to stay below 25% usage with the GPU naturally running flat out to achieve best FPS (even if that best isn't as high as some would deem tolerable). There is no discernable difference in the usage data between a high FPS mission and running the same mission when it delivers low FPS which makes it difficult to pin point any particular part of my PC that might be the cause. However as others have said it is possibly a memory thing so I'll have to start exploring that as a possible cause. I can however say it's not related to mods as running the stock game in mods -off mode still shows the same problem. HH
  8. It's not a particular mission, it seems to be any mission; just bung in any settings into the quick mission generator, fire it up, check the FPS then shut it down and fire it up again to check the FPS on the second launch, by the third or fourth time I start it I am experiencing a significant drop in FPS compared to the first time despite running exactly the same settings. I don't have an ssd just regular HD. Cheers HH
  9. To be clear, I first noticed the problem when testing the look of my mods at different times of day, but once I was aware of the problem I would test the same mission in mods-off mode with out any changes, I'd simply shut it down and immediately reload exactly the same mission, by about the third time I was reloading it was running much slower than the first time despite having absolutely identical settings. HH
  10. Thanks I'll give it a try out of curiosity to see if the sun/reflections makes a difference but unless there is some strange reason the sim retains some memory of these being used in previous missions but not other graphics settings it seems unlikely to be the cause. Principally these have been direct like for like comparisons the only thing I was changing when first noticing this problem was the time of day (which may affect the reflections by changing the position of the sun), but even that I stopped changing when testing to try identify the cause of this FPS problem. So if it is any kind of in game setting that did slow things down I'd have thought those same settings should have also slowed down the frame rate the first time I loaded each mission as this problem occurs with multiple re-loads even when I don't change any settings. Cheers HH
  11. I'm not going to derail this thread but will mention another I've just started elsewhere about an odd issue I've found that has been almost halving FPS. Solving that would be effectively doubling FPS... I will however add that on my old PC (i5-3570k 16GB DDR3, gtx970) , much like the OP the difference between blurred and sharp landscape settings is close to double the FPS. HH
  12. I don't want to derail my own thread with discussion about a strange technical issue of dwindling FPS, but in testing my mods I have encountered an odd issue affecting both mods-on and mods-off modes which is hindering my ability to reliably test any performance hit of my mods. As such I've started another thread elsewhere to see if others have encountered the same thing. HH
  13. Hi, Some of you may be aware of the mods I'm working on to try and enhance the lighting and skies in this sim. Part of this work has involved a lot of testing and optimisation to try and avoid slowing things down, where by-and-large my work in progress mods run at roughly the same speed as the stock game. My tests have, however, highlighted a strange phenomena which I spent a long time trying to attribute to my mods, but ultimately felt must be an inherent quirk of the game and tests in mods off mode have showed the same "problem" to exist and to have nothing to do with my own mods. When I first load a mission it runs fine at a satisfactory frame rate. Most people "playing" this sim would then run the mission until complete or they were shot down. However in testing and fine tuning my mods I've reason to be repeatedly loading the same mission and then reloading again to run it for only a few seconds just to check the look of things under different times of day. I've noticed the first time you load then the FPS is as would be expected but there seems to be some sort of retained memory or cache clearing problem meaning that when loading and shutting down the same mission a few times that by about the third load exactly the same scenario with identical settings is running at approx. half the FPS it did when I first loaded it just a few minutes earlier. I don't see this being a major, or even minor, problem for most users under ordinary circumstances but it is making life difficult when trying to monitor the performance hit of making minor changes to fine tune my mods. I just wondered if anybody else had encountered a similar drop in FPS when running the same mission more than once , or if they have any suggestions as to what may help clear this. Completely shutting down IL2 when I finish a mission and then going back to the game launcher to start again avoids the dwindling FPS, but it does result in spending longer opening and closing the software than actually testing settings in my mod. If this isn't just a technical issue with my own PC then for those of you who do suffer a slow running sim it may be something to think about in terms of "boosting performance" by periodically shutting down Il2 completely and re-starting the software rather than just re-launching straight back into the same mission if shot down? Cheers HH
  14. Well I keep taking small steps in the right direction... take enough of those small steps and, aside from a few occasions of going round and round in circles, I am slowly moving further and further from where I started.... Even when going back over old ground it's a chance to spot and correct earlier mistakes, as today I accidentally stumbled upon a psychedelic acid green/yellow landscape that set me off on a two hour trawl through a batch of fifty data files comprising multiple lines of code looking for a single mis-typed hexadecimal colour reference.... after failing to find this it turned out not to be the error I expected but a reference to an experimental lighting texture I'd forgotten to change back to my most recent proven texture. On another matter it is a little insane how much of a frame rate killer the above screen shots were... Previously with everything maxed out my old gtx970 was running at about 60-90FPS. The same settings on the Kuban map were at times dropping to below 20FPS, but that was the case with or without my mods and is a factor of the extra detail of the map not my mods. As I do like my eye-candy I can seriously see myself having to massively upgrade my GPU for when the WWI Arras map is released as I'll not be happy turning down the graphics settings and missing out on what these mods can do just to achieve a playable frame rate on a 5year old PC. HH
  15. Well despite having better things to do I've taken a little time this weekend to take to the hills and explore. Anybody who has ever hiked the Scottish highlands will know you have to start early and finish late whilst enduring the cold, wet, miserable and grey damp in order to get to see the spectacular views if/when the rain clears... I've no idea if the hills of Kuban are the same... but the digital ones seem to be, and as was suggested to me are a superb location to test various aspects of my sky and lighting mods as I keep working on each part and I couldn't help but think of the Highlands when the light and landscape combined on screen to look like this.. Cheers HH
  16. I've just been gifted the Kuban map and as such have been casually "sight-seeing" but I can't claim to be familiar with any of the locations to state where this happened. I was following a train through the middle of no-where heading to the coast when it suddenly disappeared and a whole series of explosions went off around me. When I paused the game to start looking around me the loco and individual wagons were widely dispersed over what must have been an area close to 1km wide and were randomly accelerating across empty ground well away from any tracks until they crashed into buildings, destroying both themselves and several of the buildings. There was no sign of anything attacking the train, and no weapons I know that would do this even if this was an "attack". I presume this must be some sort of bug? HH
  17. The damaged cityscapes look fantastic and whilst the WWII stuff does little for me this speaks really well for what we might expect from the Arras map. No-mans-land in ROF was devoid of much interest; however bombed out towns modelled to this quality along with 3D trench systems, 3D craters and much general debris and clutter would be superb. So here's hoping this is truly a suggestion of what we may see for Flying Circus...... HH
  18. Well naturally when gifted the kuban map you have to fire it up straight away and the first place you go is the mountains.... I suspect with a bit of effort to scout out the best locations for screenshots and to establish the best times to catch the sun at dramatic angles there are going to be some truly great screenshots to be had. However until then here's a case of simply hitting screen grab within 60 seconds of initially firing up the map and firstly being impressed by the "more recent" map making technology, and secondly thinking if this is what the kuban map is like I'm going to need a new PC to wring the neck of the WWI Arras map when it comes out. HH
  19. Well paying work is keeping me busy at the moment, both a restoration project on a large and valuable scale model of a Ferrari (worth more than some families full size daily drive) that the client dropped and badly damaged, plus I'm trying to get sorted for an international miniature art festival coming up in mid May. Consequently work on IL2 has been almost non existent other than a few minutes to occasionally check in here on the forum. However by the end of May I hope to have a little more spare time on my hands to get back to IL2. I also lost a fair bit of time (though thankfully not work which I'd backed up to an external HD) when windows screwed up an update and got itself stuck in an internal loop where it wouldn't complete the update to make the computer usable, nor would it allow me to jump into safe mode or BIOS upon boot up to get it to revert back to a previous install of windows.... Still it is occasionally useful to start with a fresh install of the operating system on a re-formatted HD and reinstall all the software and data files you actually still use, rather than those that are simply there because you never got around to removing them. On another topic I'm not going to mention names should it embarrass them, but I will say what a great community this is as I continue to receive private messages of support and encouragement, and this morning one such message included a gift code for the kuban map... just because they thought I'd appreciate the scenery. So whilst I've nothing new to report, I've found here another screen shot that I don't think I've posed before. Cheers HH
  20. I've been through many discussions of this nature with model makers, whether it be planes, trains or automobiles. (where at the moment I'm discussing "Ferrari red" with a client wanting a large scale model restored). Plus as someone with post-grad qualifications as a museum conservator/restorer I've spent a fair while studying, from a more academic/scientific point of view, the behaviour of paints and pigments over time. Whilst it doesn't really move the argument forward, and if anything it takes it backwards, the biggest impact on the appearance of any paint or pigment is the light source under which it is viewed. Invariably when researching anything historic we are using a variety of incomplete references from different periods to infer an answer to our question... Any photographic image used as a colour refence is little more than a "rough guide" if that photo doesn't include some sort of official standardised/certified colour chart to gauge how the light source(s) has impacted on the appearance of clearly known and defined hues/tones. With this it enables you to work back to establish the hue and tone of the "unknown" pigment from its appearance in any partuicular photo. To be honest though psychological studies have shown just how poor the human eye/brain are at judging absolute hue/tone. What matters is not getting any one hue or tone "correct" but getting the contrast between different hues and tones correct. If everything is "out" by the same amount people don't notice they just attribute it to differences in lighting. HH
  21. Thanks for the answer even if it's not the one I wanted to hear. I had suspected as much, but I'll report back if I find a creative way around this. HH
  22. Whilst I really have better things to be doing with my time I've got little distracted with a "tangential" project which has prompted the following question; Is there any way to implement a bespoke normal map and associate it just with one specific skin, or can you only edit that aircraft type's regular normal map and hence in doing so change each and every example of that aircraft type? Cheers HH
  23. Whilst I really have better things to be doing with my time I've got little distracted with a "tangential" project which has prompted the following question; Is there any way to implement a bespoke normal map and associate it just with one specific skin, or can you only edit that aircraft type's regular normal map and hence in doing so change each and every example of that aircraft type? Cheers HH EDIT: Oops! apologies as I've just realised I've posted this in the screenshots section of the forum rather than the more appropriate skins section.
  24. it is the common textures folder data/graphics/textures/common/ac_common.dds I'm not normally one to be overly critical of the devs work but it is a somewhat ugly bit of texture work, which is best left as very matt black. Lightening it and tweaking the alpha channel show all it's faults. I'm basically started again with it by taking the larger texture for the Vickers MG and resizing and positioning the different bits of it to match the layout of the common texture. So, what should have been a quick 30second fix has become somewhat more but the WIP shot shows that you can make a visual improvement. HH
  • Create New...