Hotfix 4 Final Released!
#16
Posted 31 August 2010 - 03:17 PM
#17
Posted 31 August 2010 - 04:12 PM
As for the stuttering, it seems partially fixed since hotfix beta 3. In gpfce for example, it's not stuttering anymore. Mame4all however still seems to stutter.
MAME4ALL has sound problems, those are not related to any Hotfix
#18
Posted 31 August 2010 - 05:04 PM
This sounds even better. I like that some of the User-made Tweaks will be included. Comunity Power!New things planned for HF5:
* Tweak settings dialogue (Select default filter, include some of the tweaks found here in software hacking, etc.)
* Remove caching to NAND for as many programs as possible
* Include OVR-Editing in MiniMenu![]()
And because I'm also not a fan of apps that write stuff somewhere to the NAND, I welcome the plan to "clean" programs in this case. Otherwise I can imagine that the NAND will be full some day and you don't even know what causes this.
#19
Posted 31 August 2010 - 06:34 PM
#20
Posted 31 August 2010 - 07:36 PM
e.g.
sudo date --set=20:32:00
It doesn't update immediately, but a restart or switching GUI afterwards certainly works.
#21
Posted 31 August 2010 - 07:54 PM
As for the stuttering, it seems partially fixed since hotfix beta 3. In gpfce for example, it's not stuttering anymore. Mame4all however still seems to stutter.![]()
MAME4ALL has sound problems, those are not related to any Hotfix
I am certainly no expert, but mame4all did not stutter before this hotfix, and now it stutters even with sound emulation disabled entirely. I'll survive, just sayin'.
#22
Posted 31 August 2010 - 08:21 PM
#23
Posted 31 August 2010 - 08:30 PM
(thats the 'speed on leaving mmenu' setting.)
(in beta 4, I removed the 'speed inside mmenu', since it seemed to crash when reading SDs at high speed (!!) during image caching, and added the toggle for 'change cpu speed on leaving' (yes/no, default no). So by default it should do nothing it didn't already do, speedwise. (Even if you'd used beta 3 and had the speed mucking, going to beta 4 would have disabled that.)
Mind you, going to the new hotfix using mmenu, would probably show more tabs than folks had seen before, so that they could then turn them off using the UI. So I have worried that people had only a few tabs, now have like 20 tabs, and don't notice they can disable them now.. maybe I should put a 'random tip' into the detail-box on startup; I did add 'hit select for the menu' or whatever there, as a tip off, though..
jeff
#24
Posted 31 August 2010 - 09:32 PM
Any suggestions, please?
The good news is that this update, inexplicably, fixed the problem I had been having with NesEmu failing to save states properly, and failing to run ever again after trying to save a state. I have no idea why this would fix that issue, but it did. Thanks.
EDIT: Oh, and I must say that I absolutely LOVE the speed setting dialogue. It's so useful, as I often forget to manually adjust my speed before running something that needs an overclock. Thanks very much for that one - it's an excellent addition and a great bit of polish.
#25
Posted 31 August 2010 - 09:36 PM
I guess I'll give you guys a break from my valuable feedback.
#26
Posted 31 August 2010 - 09:51 PM
ED, what about my UFT8 proposal, is it still in the testing phase?
Yep, if you browse through the internet, you can find a lot of problem posts regarding UTF8 with VFAT - definately not something I want to put into the final without real testing.
I need to get to a Windows Box, too, and create files with Umlauts first
#27
Posted 31 August 2010 - 10:41 PM
#28
Posted 01 September 2010 - 02:16 AM
Cool now if I could just click on an icon through minimenu to toggle that would be cool. Don't you have to enter passwords and stuff to get it to work with that command line method? I hope apps start adding this as an option in their menus in the future, would be convenient.The FIR filter thing is really nice, you can disable the blurry scaling in GINGE etc. and use nearest neighbour.
All you have to do is run: sudo /usr/pandora/scripts/op_videofir.sh none
![]()
Caching to NAND? That sounds like a really bad, bad thing. When the NAND gets full and/or if it gets corrupted you have problems, or just wears out you have a brick.New things planned for HF5:
* Remove caching to NAND for as many programs as possible
Why are apps still doing this? I thought this appdata thing was supposed to keep apps from doing that? They should write all of their crap there. There should be some kind of rule that if you don't absolutely HAVE to write to NAND you shouldn't. Also I think apps that do should warn you in their documentations that they do so you can choose not to use them.
Also there should be a guide of how to delete all of the useless (to some) shit like demos, unwanted apps, card games etc.
Edited by DaveC, 01 September 2010 - 02:27 AM.
#29
Posted 01 September 2010 - 03:21 AM
ED is working on the filter selector so should be in HF5 but yeah, apps will have to look into it perhaps, and maybe a libpnd change to force (?) it back to default on exit so misbehaved apps dont chabge it and forget to change it back?
Should also add in an option so appdata can group with the app, for you
Only a couple apps are spitting to nand, like chromium .. Not really ported to pandora, just hacked to run on it, so doesnt follow all the rules, but likely an easy workaround
Jeffphone
#30
Posted 01 September 2010 - 03:23 AM
PNDs usually don't use cache on the NAND.











