Seriously, guyz???
-
I'm tired of useless mouse drivers which replicate existing settings dialogs just to put the vendor's logo into the dialog, constantly consume 5% of your CPU time, add toolbars to all of your browsers and manage to break both your AV software and your disc burning software in the process.
those mice that i find that do that have two fates in store for them
- they are thrown against a concrete wall or driven over by a car until they are broken beyond recognition.
- they are then carefylly resealed in their packaging without disturbing the tamper evident seals that prevent unscrupulous people from doing what i'm about to do in the third step 1
- the mouse is RMA'd direct to manufacturer as DOA. for cash refund not replacement. :-D
-
it's mapping the custom actions to them that doesn't
I don't think that's what's happening. I'm pretty sure that most applications map additional mouse buttons to actions like Forward and Back by default by now, I just need to be able to tell Windows that Mouse Button #4 is actually Mouse Button #5 and vice-versa.
all the standard HID stiff (what's baked into the kernel/X11 in linux) works out of the box, but the fancy mouse specific settings don't (neither do they in Linux either so at least that's parity)
So thumb buttons are a "fancy mouse specific setting"? I'd say they're pretty standard by now and have been for years.
-
So thumb buttons are a "fancy mouse specific setting"? I'd say they're pretty standard by now and have been for years.
let's agree to disagree on that one. mkay?
-
So thumb buttons are a "fancy mouse specific setting"? I'd say they're pretty standard by now and have been for years.
Go to Best Buy and look at the available options there. I bet you that at least 75% don't have thumb buttons.
-
let's agree to disagree on that one. mkay?
I literally don't remember the last time used a mouse without back/forward thumb buttons. The one I'm using at work right now (which exhibits the problem) is a cheap, standard Fujitsu mouse which came with the computer, nothing fancy.
Go to Best Buy and look at the available options there. I bet you that at least 75% don't have thumb buttons.
You may be right, but I'm pretty sure you'll also find at least one cheap mouse with back/forward thumb buttons.
-
You may be right, but I'm pretty sure you'll also find at least one cheap mouse with back/forward thumb buttons.
That's still not "pretty standard".
-
That's still not "pretty standard".
But not a "fancy mouse-specific feature" either. Common enough that although I didn't install the mouse driver (because it doesn't work on Windows 8), most applications seem to recognize those buttons somehow (just not correctly).
-
I think by default Windows recognizes #4 and #5, but nothing further. I've got an 11-button WoW mouse* and the only buttons that work without binding (using the crappy configuration software) to keyboard actions are those two.
* I've never played WoW, but when I was building my new computer there was a huge sale on the mouse, and $18 for a $90+ Steelseries mouse is a great deal, even if it has an ugly glowing logo.
-
But not a "fancy mouse-specific feature" either. Common enough that although I didn't install the mouse driver (because it doesn't work on Windows 8), most applications seem to recognize those buttons somehow (just not correctly).
No one in this topic has ever said that back and forward thumb buttons are a "fancy mouse-specific feature". In fact, a quick search through the thread shows this as the first time that "thumb" even showed up in the topic:
For some reason Windows 8 decided to switch the thumb buttons of my "new" mouse
The conversation drifted a little from there until @accalia said something about standard HID stuff working out of the box with linux, but not "fancy mouse specific settings". You then reached back and decided that she meant that thumb buttons are one of those fancy settings here:
So thumb buttons are a "fancy mouse specific setting"?
You were the first to make that specific connection. The argument is between you and your shoulder aliens.
I was only challenging your assertion that they are pretty standard.
Filed Under: Is there a @shoulderaliens? No?
-
No one in this topic has ever said that back and forward thumb buttons are a "fancy mouse-specific feature".
I quote:
asdf:
So thumb buttons are a "fancy mouse specific setting"? I'd say they're pretty standard by now and have been for years.let's agree to disagree on that one. mkay?
That sounded like she meant to say that thumb buttons are a fancy mouse-specific thing.
The argument is between you and your shoulder aliens.
Hm, yeah, I might have battled a straw man without noticing. Is there a badge for that?
-
That sounded like she meant to say that thumb buttons are a fancy mouse-specific thing.
more that the fact that they do anything more than generate a "
Hey! Someone pressed mouse button 42!" event was a fancy mouse feature.
-
more that the fact that they do anything more than generate a "Hey! Someone pressed mouse button 42!" event was a fancy mouse feature.
In this case, we don't disagree.
-
hmm... really? daang i read your posts the wrong way around then. Sorry guv'ner.
-
hmm... really? daang i read your posts the wrong way around then. Sorry guv'ner.
We both did ;)
Just to clarify: I don't expect my thumb buttons to launch North Korean nuclear missiles, I just want a way to tell Windows that button #4 is actually button #5 and vice-versa so that applications don't do stupid things when I click those buttons.
-
Best buy may have the Corsair I bought, which is lovely. It's thumb button halfs sensitivity for sniping. I haven't seen this other kind.
-
I just want a way to tell Windows that button #4 is actually button #5 and vice-versa
Screw it. Open up the mouse and see if you can switch a couple of wires.
-
Is it worth the expensive salary for someone who knows the Linux environment well enough?
GitHub - Onyx47/chan_extra: Updated version of chan_extra driver fo OpenVox G400P/G400 for kernel 3.10+ and Asterisk 11
Updated version of chan_extra driver fo OpenVox G400P/G400 for kernel 3.10+ and Asterisk 11 - Onyx47/chan_extra
Took me 2 hours of research, a trip to the docs and vim. Yes, it only took 2 hours, even though I used vim, which I'm crap with!
It now successfully runs on three different machines, never crashed. Had no prior experience with Linux kernel or driver development. Oh, and most of the problem wasn't even moving it from 2.6.x kernel architecture to 3.x, it was another dependency with Asterisk.
While writing the driver from scratch would be a daunting task requiring much more learning, my limited anecdotal evidence would seem to point to the fact that maintenance is not that bad. The caveat listed in the description is not the kernel changing btw, it's libdahdi changes, which is used by... maybe 10ish device models, overall? And is specific to Asterisk, not some trivial daily use.
-
Right; but there's the costs of finding someone like you (not trivial), the whole nasty situation where they need you for like a couple weeks, then don't need you for a random period of time X, which is unknown, which means they need to come up with some sort of retainer contract. Or the alternative is to hire you full-time, but 90% of the time you'd be spending doing something that isn't your expertise, or plain nothing at all...
Look, you figured it out in 2 hours. Whoop-de-doo. That doesn't change the fact that it's a huge investment for the hardware company, one most smaller companies can't afford. The equation still applies.
If the Linux community wanted to change the equation, they could set up something like a single official "driver making contract agency" which would already have handled all the contract work, and a client could simply hire them for a flat fee to create the driver and maintain it for X years. In short, make Linux driver development easier and you'll get more Linux drivers.
-
That doesn't change the fact that it's a huge investment for the hardware company, one most smaller companies can't afford.
I never contested that point. Just the bit about the development being very hard.
Does it cost money? Well of course it does. But so does paying someone to write Windows drivers. It's up to the company to decide if the cost is worthwhile. And they do have a slight advantage of not paying a dime if they release the specs required since there are many people who will willingly contribute code.
-
And they do have a slight advantage of not paying a dime if they release the specs required since there are many people who will willingly contribute code.
I don't know about you, but if I made hardware, I'd never do that. To be completely out-of-control of the driver for your device? You device could get a lousy reputation for being a piece of crap just because the only "willingly contributing code" guy was an idiot and wrote buggy, incomplete code?
No way, I wouldn't want the risk. Better to ignore the platform altogether.
-
No way, I wouldn't want the risk. Better to ignore the platform altogether.
So... Linux is a hostile platform for driver developers because driver developers want to get paid? Windows driver devs do it for free?
-
What? No. What are you talking about?
Did you just suddenly get amnesia about the original of this discussion or... what's going on?
To repeat myself from yesterday, adding made-up numbers: the difference is your Windows driver development costs $20,000 and guarantees $2,000,000 in revenue. The Linux driver development costs $50,000 and guarantees $??? in revenue.
The point I was actually making in the post you were replying to is leaving driver development to some random dude who uses Linux could easily result in a terrible driver that sullies your product's reputation, and now suddenly the Linux driver produces negative $??? in revenue.
-
Did you just suddenly get amnesia about the original of this discussion or... what's going on?
I somehow mangled two posts in my brain while reading, multitasking, sorry.
Again, not saying you're wrong on that cost / benefit and that's for every manufacturer to decide after crunching numbers.
The point I was actually making in the post you were replying to is leaving driver development to some random dude who uses Linux could easily result in a terrible driver that sullies your product's reputation, and now suddenly the Linux driver produces negative $??? in revenue.
To be honest, if your hardware is popular and available enough, it's going to happen. See: nouveau and radeon. AFAIK, they are both reverse-engineered (and actually work well enough to run a composited desktop, but not much more). Releasing enough of the spec for someone to write a driver is actually advantageous in that case. It might not be the best driver ever, but it sure will be better than the crap you'll get otherwise.
-
they are both reverse-engineered
Yeah, and the company who makes the hardware could sue over it if it damages their reputation because they in no way condoned the making of the driver. Really, their best interest is to make it non-obviously easy to reverse engineer and go on record saying that they do not approve any third party drivers.
-
Yeah, and the company who makes the hardware could sue over it if it damages their reputation because they in no way condoned the making of the driver.
Maybe, IANAL so I won't comment on that. I'm just reporting the facts here.
As for the second part (which I would quote but Discourse is being an ass), I'd assume that would bite then in the ass as well. It's that cost / benefit calculation once again, and I won't pretend I'm smart enough to preach about that to anyone.
-
Apple controls both sides of the ball in that case, so they can pretty much well test the living daylights out of their drivers, at least for important parts.
Testing unimportant parts, of course, is left to third parties.
-
The Linux driver development costs [much more than the windows development]
[citation needed]
-
I don't know about you, but if I made hardware, I'd never do that. To be completely out-of-control of the driver for your device? You device could get a lousy reputation for being a piece of crap just because the only "willingly contributing code" guy was an idiot and wrote buggy, incomplete code?
In my (firsthand -- I have had driver blowups on Linux before, namely the really aggravating "b43 FATAL DMA ERROR") experience, if the driver's a buggy piece of crap, the driver will get the blame -- the Linux world, as a rule, isn't one to just pin blame on a scapegoat and walk away. Besides, sooner or later, someone will get fed up with the buggy driver and either fix or rework it -- remember that the people who write drivers for quite a bit of hardware on Linux are writing drivers for hardware they have and use themselves.
. AFAIK, they are both reverse-engineered (and actually work well enough to run a composited desktop, but not much more).
Nouveau is indeed REd; Radeon is written at least partially from specs that ATI/AMD has released over the past couple of years or so, though. The issue with graphics card drivers is mostly that 'renders correctly, but only gives me 2fps' is where most of the open source drivers are at -- FWIH, radeon's good enough to drive most ATI/AMD cards for uncomplicated games.Releasing enough of the spec for someone to write a driver is actually advantageous in that case. It might not be the best driver ever, but it sure will be better than the crap you'll get otherwise.
Yes -- releasing specs is generally a good idea for chipmakers; it means that you'll get people wanting to use your chip for wild and wonderful things you and your initial customer base would have never dreamed up!Yeah, and the company who makes the hardware could sue over it if it damages their reputation because they in no way condoned the making of the driver. Really, their best interest is to make it non-obviously easy to reverse engineer and go on record saying that they do not approve any third party drivers.
That is, if you want to alienate customers -- suing your own customers is never in your best interest. Why would you sue over buggy code instead of fixing it when the latter option is well within your power as a hardware vendor?
-
In my (firsthand -- I have had driver blowups on Linux before, namely the really aggravating "b43 FATAL DMA ERROR") experience, if the driver's a buggy piece of crap, the driver will get the blame -- the Linux world, as a rule, isn't one to just pin blame on a scapegoat and walk away.
+1. Anecdotal, but that's what I got as well.
Speaking of b43, Broadcom found a way to have the "secret sauce" bit without really writing much code (presumably) - the driver is in the kernel, but requires a binary firmware package from Broadcom.
The issue with graphics card drivers is mostly that 'renders correctly, but only gives me 2fps' is where most of the open source drivers are at -- FWIH, radeon's good enough to drive most ATI/AMD cards for uncomplicated games.
It's pretty good, actually. It gets slightly lower FPS, but DPM actually WORKS.
fglrx(AMD's official driver) can only handle "off" and "melt your laptop" settings. I have to switch to the AMD driver and then reboot (no, can't domodprobe fglrx, the module refuses to get dynamically loaded). Honestly, the only reason I have it installed is because some games refused to detect my GPU as supported without it.And no, I did not but a laptop with AMD GPU willingly.
-
I just want a way to tell Windows that button #4 is actually button #5 and vice-versa
X-Mouse Button Control can do this. It also lets you scroll the window under the cursor instead of the active window.
-
X-Mouse Button Control can do this. It also lets you scroll the window under the cursor instead of the active window.
Thanks, I'll try it out on monday.
-
Yes -- releasing specs is generally a good idea for chipmakers; it means that you'll get people wanting to use your chip for wild and wonderful things you and your initial customer base would have never dreamed up!
QFT. This is actually a significant part of my employer's marketing strategy. Random customer uses one of our chips for something we never thought of. Wow, great idea! Start selling chips to other customers that make that kind of widget.
-
-
Apple controls both sides of the ball in that case, so they can pretty much well test the living daylights out of their drivers, at least for important parts.
As to add-on USB doodads and such? Correct me if I'm wrong, but I think libusb works on OS X...
The answer is badly. Fortunately most Mac users don't use anything that isn't standard.
For example: good luck getting a USB-wired XBox controller to work on OS X. There is a third party driver for it but support in games for it is terrible at best.
It's actually better in my experience to plug such a thing in, and run games in Parallels than it is to use OS X native games for such.
-
To be completely out-of-control of the driver for your device? You device could get a lousy reputation for being a piece of crap just because the only "willingly contributing code" guy was an idiot and wrote buggy, incomplete code?
My kids use my old laptops. My daughter has my most recent one, which came with Win7 on it. It's a Core 2 Duo with a Radeon HD 46xx (whatever on it). She upgraded to Win8 last spring. Then 8.1 this summer. Everything seems fine (aside from being stuck on Win8, of course), except Minecraft now refuses to run on her machine. Apparently, the drivers for it are all old and don't support newer versions of something or other in windows. And apparently Minecraft thinks this is important and now refuses to work for her.
My son is on a much older machine. Not sure what the CPU is, but it's some sort of dual core thing with, like 1.6GHz speeds and Intel graphics. It's been running Kubuntu forever (currently on 14.04). Minecraft still runs just fine there. It's not blazing fast or anything, of course, but it fucking works and he has fun with it.
Fuck you Windows for your shitty hardware support. That's maybe a bad assignment of blame, but it's accurate as far as what changed and broke stuff that used to work.
-
The blame goes to AMD. Everyone knows they don't know how to write drivers. Making fast graphics cards is fairly worthless when your drivers have a 1/10 failure rate regardless of operating system.
Intel graphics may be weak, but they at least work.
-
Yeah I was gonna say there's two parties to blame in your Cool Story, Bro, and neither of them are Microsoft.
What kind of unmitigated gall do you have to have to write a game engine in Java then deny a user running it due to lacking some obscure feature? Goddamned, Mojang.
-
Well, to be fair, Minecraft is now owned by Microsoft.
-
Ok; on a technicality one of the two parties is Microsoft. Feh.
-
Yeah I was gonna say there's two parties to blame in your Cool Story, Bro, and neither of them are Microsoft.
Well, it was their one thing that changed.
What kind of unmitigated gall do you have to have to write a game engine in Java then deny a user running it due to lacking some obscure feature?
Honestly, that's all I've been able to figure out from reading random forum stuff, was some guy talking about WDDM 1.2 or whatever. And I looked and her display property control panel thingy had 1.1 buried somewhere in the guts. I don't have enough knowledge of what any of that is to know if there's something legitimate in there, but something MS updated broke stuff.
And that's the sort of thing the Linux model of putting stuff into the kernel tends to prevent.
-
Fuck you Windows for your shitty hardware support.
It's more likely AMD stopped updating drivers for that generation of card. You can probably find "legacy" drivers on their support page.
Plus, to make things worse, some companies like Dell used to make AMD/NVidia block the drivers from working on their brand computers to force you to use the company's branded drivers. But at least with that you could find 3rd-party programs that would undo the block.
You might be able to get drivers from Windows Update.
-
People with nvidia and even Intel cards don't have this problem. Besides, AMD also hates OpenGL and has only recently become viable on Linux. They cannot be held blameless.
-
It's more likely AMD stopped updating drivers for that generation of card.
Yes, they did. It's not like the machine got thrown back in VESA mode or something. Everything else seems fine.
You might be able to get drivers from Windows Update.
Nope. I think there's a way to select a non-WDDM driver, but I haven't tried it.
People with nvidia and Intel cards don't have this problem.
Neither do people with Linux (they have different problems, obviously).
-
lacking some obscure feature
"Accelerated pixel format" is hardly an obscure feature. But I don't know, maybe you like running games with sub-integral fps.
-
Well, it was their one thing that changed.
Only if the Radeon drivers didn't change version during the upgrade, which is far from clear.
And I looked and her display property control panel thingy had 1.1 buried somewhere in the guts. I don't have enough knowledge of what any of that is to know if there's something legitimate in there, but something MS updated broke stuff.
I don't see any evidence Microsoft broke anything.
Yes, they did. It's not like the machine got thrown back in VESA mode or something. Everything else seems fine.
That's why I'm saying Mojang has a lot of fucking gall. They use the most bloated runtime environment possible to build a game engine, then get pissy when someone's driver isn't up-to-the-second? Hypocrites.
"Accelerated pixel format"
First of all, that's gibberish.
Secondly, assuming it's not gibberish, you somehow got there from here?:
Apparently, the drivers for it are all old and don't support newer versions of something or other in windows. And apparently Minecraft thinks this is important and now refuses to work for her.
I am staggered by your telepathy.
But I don't know, maybe you like running games with sub-integral fps.
Yeah well like I said, they already picked Java for the damned thing. It's not like anything AMD does at this point could make the game slower.
-
"Accelerated pixel format" is hardly an obscure feature. But I don't know, maybe you like running games with sub-integral fps.
It always worked fine in the past, so the existing driver / hardware is sufficient. It's apparently something about the way Windows works with the driver. I mean, that's my interpretation of the whole thing.
What's also interesting is people talking about MC stopping working at Win8, but I know she was playing since then, but I'm not sure that she was since upgrading to 8.1.
-
That's why I'm saying Mojang has a lot of fucking gall. They use the most bloated runtime environment possible to build a game engine, then get pissy when someone's driver isn't up-to-the-second? Hypocrites.
Yes, honestly, I think Mojang deserves the blame here, but you'll note I talked about who gets the reputation. It's especially frustrating given how hard MS works on backwards compatibility.
It's not like anything AMD does at this point could make the game slower.
Well, it was fast enough.
Only if the Radeon drivers didn't change version during the upgrade, which is far from clear.
Possibly. The date on the driver is June 2012. It's possible that 7 and 8 were using even older drivers. That would be another WTF, of course.
-
It could also be the result of driver consolidation. Maybe before you had a driver specific to that exact card, from the PC's maker, and after the 8.1 upgrade you have the "generic driver for all Radeon cards made between 2005 and 2013". And the latter has different compatibility flags from the former.
You could try updating the driver from the PC maker's site, but since it shipped with 7, they might not have a Windows 8 version of it available. (Which may or may not matter.)
-
It always worked fine in the past, so the existing driver / hardware is sufficient. It's apparently something about the way Windows works with the driver. I mean, that's my interpretation of the whole thing.
From the data points you've given -- it sounds like you're right: the Windows team diddled something in how Windows talks to display drivers (WDDM in other words), and ATI/AMD was a bunch of dummies who depended on some behavior that wasn't in the contract, or was changed when WDDM was upgraded. Considering that GPU vendors don't keep drivers for old GPUs up-to-date...I'd say the blame gets split between Mojang and ATI, although of course, Microsoft would catch the flak from a non-techie anyway.
The problem with telling a non-techie "it's a driver problem" is that some of them are too think to wrap their heads around it, and some of those folks have lots of lawyers and deep pockets...
-
The problem with telling a non-techie "it's a driver problem" is that some of them are too think to wrap their heads around it, and some of those folks have lots of lawyers and deep pockets
Minecraft actually says something like, "Your drivers are out of date," so it's obviously something to do with drivers. I tried installing the latest version of Catalyst nonsense, and that failed for some reason.
I may just throw a VM on there and be done with it.
