Jason Carr

It will only start repopulating the cache once you're back in LaunchBox and accessing more images.

@dukeemu Since we haven't yet translated any of the release notes stuff for the Italian translation, there's a bit more to translate there. Feel free to translate this stuff if you have time before Wednesday; otherwise it can wait for the next release. Thanks. Welcome to LaunchBox RELEASEVERSION! Here are the biggest changes for this release: Sound packs and separate volume controls for are now available for Big Box The LaunchBox XML file has been divided up; check the Data folder inside of your LaunchBox folder Many new performance improvements for both LaunchBox and Big Box See the changelog under the Help menu for more. If you appreciate LaunchBox, please Like us on Facebook and help us spread the word! THIS IS AN UNOFFICIAL BETA RELEASE. Please do not distribute this version of LaunchBox, as it is not an official release and could be broken. See the changelog under the Help menu for further details on the changes for this release. Please report all issues that you run into using the Report Issue option under the Help menu. Thanks for using LaunchBox! To get started, you can: Install or Import MS-DOS games Import your games from Steam Import your ROM collection See what LaunchBox can do View the changelog under the Help menu for details on the latest version. Support LaunchBox!

Thanks @sfn103; I'll get a new translation package out soon. I'll just use ar-SA. @Opaklopper Just put out a new beta with your updated translation. So Retropalooza is coming up here soon guys and I'm leaving early Thursday morning for a week. So I'm planning on putting out the 6.10 official version early Wednesday morning so that there's a safety period in case something goes wrong. My apologies for the short timeframe there. There's not a lot of translation changes for this version at least; I think the Strings file is the only one that changed. If you guys are able to get the translations in before Wednesday morning, that would be awesome. If not, there's not that many changes so it shouldn't be a huge deal. It's always appreciate so much guys, so thank you. In addition, here's the release notes for 6.10 that can use translating: Sound packs and separate volume controls for are now available for Big Box The LaunchBox XML file has been divided up; check the Data folder inside of your LaunchBox folder Many new performance improvements for both LaunchBox and Big Box Thanks again all.

Yeah, definitely. That's the way it's supposed to work. Thanks @StuDentBR; perhaps this is a bug that was introduced in a recent version. I will look into this tomorrow.

Alright, I've put out a couple new betas today guys that have mostly minor bugfixes, but I've also majorly restructured the caching and image rendering for LaunchBox itself (again). Scrolling performance should be much better, in theory. More details are here: Would really appreciate any testing you guys can do in LaunchBox scrolling and switching between filters, etc. It makes me nervous to have made these changes so close to Retropalooza, as I need to put out an official release before I leave on Thursday, and ideally I need a period after the release to fix stuff if something goes wrong. At this point I'm hoping to put out an official release on Wednesday morning.

Thanks all. @Maddoc1007, I think we ended up settling on the no background videos for the fullscreen coverflow view for performance reasons. There's not much that shows up in the background anyways for that view so we moved on from it I think. I also tested the PlatformWheelVideoDetailsThumbsFiltersView thing and I was able to get the clear logos to show up just fine from the theme folder. It is confusing though because you have to have the right theme and refresh the images in order to get it show up. I did just add some extra code in the live stream to automatically refresh the platform wheels to make it less confusing. @shinra358 I will look into the mouse click issue there. @ckp I haven't been able to fix the LaunchBox VLC thing. I think it's a problem with the VLC interop library (different one than we use for Big Box that we just spent three years working on) and it's gonna be a royal pain to fix. At this point I'd say switch to WMP in LaunchBox if you can. @martinconroy Thanks for creating that thread. Really hope we can figure something out.

Alright, so I ended up spending some significant time on this this afternoon. I made some major structural changes so whatever you guys can do to test, that would be much appreciated. The changes are of course in the latest beta that I just put out (6.10-beta-8). With these changes, in theory the scrolling should no longer stutter. You will probably see more popping in now that the scrolling is not stuttering while it's trying to load the images from disk. If you're seeing too much popping in, you can increase the RAM cache. On a related note, you can no longer completely disable the RAM cache. I was able to optimize things quite a bit in the process by getting rid of that option. You can of course still set the RAM cache to a very low value, and check the box to clear the cache when launching games, so it should be mostly a non-issue having that option removed, as you can just set it to a low value. Thanks guys, please test and let me know how it's working. I need to put out an official version on or before Wednesday, so hopefully I didn't just break things.

Hi @Walrus13, good find. I'm seeing this in Big Box as well. About to fix.

Looking incredible @viking.

@Imgema It is somewhat normal for the scrolling to stutter a bit every once in a while, but that usually happens during caching, not afterwards. Another thing that's worth noting is that scrolling performance does go down slightly if you have the boxes sized smaller, because more boxes have to be rendered on the screen at a time. Now that I'm understanding the issue better, though (I completely didn't understand that it was scrolling performance for some reason), I have an idea as to what's going on. Previously we were actually caching things in the background to help improve the scrolling performance, but I tried taking that out to see if it helped with some other things. What you're seeing is the typical behavior for when items are directly loaded from the disk cache and not from the RAM cache (which is configured in the Options, of course). In general I see much better performance in that scenario, but I'm pretty sure that's what's going on. I'll try enabling the background caching again to see if that helps. That will basically do it's best to automatically cache things into RAM in the background to prevent the scrolling delays.

Yes, check the emulator's application path and make sure that it's accessible. Most likely the drive just got reconfigured or went missing for a bit.

Thanks @SpaceMidget75; that's great to hear. Also it's always great to have other developers here in the community. @timekills It's been a while but I've used LAMP quite a bit as well.

Hi @Kagenryu, just FYI, I did fix this in the latest beta.

Within the next couple of days we plan to release the 6.10 official version, actually. The XML changes are stable at this point in the latest beta. I don't foresee any XML structure changes from here on out, at least not for a long while. The XML structure has been pretty uniform since the beginning, sans for adding fields here and there.

Wow, that was a bit unexpected; I'm sorry to hear. I thought you were on board @Antropus after the discussion in the beta thread. We've focused quite a bit on performance in the latest betas for what it's worth, but there's always more that can be done, and we will continue to focus on performance improvements. Bloat is a real thing but I do believe we've done a good job combating bloat and choosing only the best features to implement, but I guess that is a matter of opinion. Per the XML changes, it's worth noting that literally nothing changed sans for separating platforms and settings out into individual files; the XML structures stayed completely the same, so to be honest the code changes that would be required should be fairly minimal. But I can certainly understand how life gets in the way of side projects. Thanks for updating us @Antropus.

Currently there's no way to disable or move the hard drive cache. It's really a non-issue since the files are only regenerated if you change the images, so it shouldn't be a stress on any hardware. That said, @Imgema is running into some issues that we're addressing in another thread, so I'm guessing that's the reason.

@Imgema That's pretty odd, and that's the first time I'm hearing of that issue. It's quite strange for any app to behave differently after a reboot than otherwise, so there must be something odd going on. I'm really not sure how we would troubleshoot that one though. What kind of lag are you talking about, and how long is it for? It's also really strange for it be slower after the images have been cached.

@kerszr What version are you using? The betas or 6.9?

Great to hear.

I can't explain why it would have been working before the update, to be honest, other than maybe game controllers got switched on or something. The xinput library is used for game controllers. It's not necessary to run the setup on the other computer as well, only on the one that is having issues. Also, if you're worried about not updating your LaunchBox installation, you can just install to a different folder and then delete it.

Oh, also, no, reinstalling will never overwrite any games or settings.

This is happening specifically because LaunchBox is synced to different computers without installation, which is normal, but unfortunately certain DirectX libraries are required. Users get this message when those DirectX libraries are missing. LaunchBox is not causing DirectX issues; it's simply saying it needs some missing libraries. Re-running the setup program on the computer should fix the issue.

Thanks @Opaklopper; I'll get the update into a beta on Monday probably. @sfn103, that's interesting. Which do you know the best? I'm surprised there's no option for Classical Arabic in the .NET language codes.

Yeah, it didn't really make much sense because the texture was exactly the same size as the object it was being painted on. But ultimately I just had to make the texture a pixel or two larger than the object, which was a trick in itself since I first had to figure out how to stop it from automatically resizing the texture to match the object. Basically we were letting .NET handle too much without explicitly declaring enough of the rendering logic. I still think it was a .NET glitch in that the previous approach should have worked just fine, but thankfully I was able to get around it with some explicit code.

It was something like a math rounding error with the rendering of the wheel. It was actually a real pain to figure out and fix, and looking at the math it *shouldn't* have been an issue. But I ultimately had to fudge the math behind the painting to overlap the drawing by a pixel or two in order to prevent it.