Well, if HTC won't do it, someone needs to... I want to see if it can be done. Can we create our own ATi Imageon Drivers for the MSM7200/7500?
step 1: Find out which Imageon is in there...
Here's the official list
I just hope it's not some custom job!
Think it's worth a go?
^^Bump^^
This definitely needs to be done. I wish I had the know how to develop this but unfortunately I don't.
you should really try posting this somewhere else, like maybe in the development and hacking forum, since the vogue is not the only device that suffers from this.
this is really a great idea in my opinion, and im sure there are those out there that know how to get it done!
Would be awesome, but it seems like it would be a huge undertaking. I'd imagine that Qualcomm has it locked in some way, preventing it's use if the license to use that component wasn't purchased as part of the hardware platform. I really have no idea though. I'll keep hoping : )
A journey of a thousand miles begins with one step... we have a ton of talent here, HTC and Qualcomm don't seem to want to release it... I think that it might be time to start building our own.
Post copied to development and hacking but I quoted first post of this thread to give proper credit for the idea.
I remember reading somewhere that there is actually a $2000 (and growing) cash prize for the msm 7xxx drivers. Totally support the idea.
yep... I think it's up to 7 grand now.
I'm going to wait until the end of March before I start anything.
Draiko said:
yep... I think it's up to 7 grand now.
I'm going to wait until the end of March before I start anything.
Click to expand...
Click to collapse
well if anyone could point me to link i willing to offer 200 to play quake on my phone
problem is that writing device drivers is the hardest program one can get into
the driver is limited to a very tiny mem footprint and must never be alowed to crash
even if the memory it's using is lost on the heap
and one needs access to alot of whitepapirs from the hardware maker
which is prob why htc did such a poor job writing then driver themselfs
because they were too cheap to pay qualcomm for it
Rudegar said:
problem is that writing device drivers is the hardest program one can get into
the driver is limited to a very tiny mem footprint and must never be alowed to crash
even if the memory it's using is lost on the heap
and one needs access to alot of whitepapirs from the hardware maker
which is prob why htc did such a poor job writing then driver themselfs
because they were too cheap to pay qualcomm for it
Click to expand...
Click to collapse
It's not supposed to rain on parade day...
I used to write assembly language device drivers for embedded real-time systems. Doing so requires a high level of knowledge, experience and documentation WRT:
1. Dev & testing tools and environment
2. The target hardware (detailed doc)
3. The target OS kernel and driver hooks
4. System-level programming mind-set (different from app dev)
Quite interesting and rewarding if you have the above - quite difficult, bordering on 'not gonna happen' if you don't.
Sincere best wishes to those who give it a shot!
My entire impetus for this thread was to start gathering people with the skills and talent needed to tackle driver construction. If open-source maniacs can do it, we can too. I have some experience with assembly... the tricky part is going to be finding out the specific commands, paths, and function calls for the embedded hardware. There's going to be a LOT of trial and error involved.
I'm starting to look for the Windows CE DDK right now. Come April 1st, if there is no word from HTC, Qualcomm, or Sprint on this driver issue, I'm going to round up as many capable people as I can and try to hammer this out.
Something that might shed some more light on the entire driver issue... Uh oh
Big Uh oh...
if this ever comes together fully it will be really cool but i still dont get why SE didnt include drivers in the first place??
comeradealexi said:
if this ever comes together fully it will be really cool but i still dont get why SE didnt include drivers in the first place??
Click to expand...
Click to collapse
As posted before, they probably bought it at a cheaper price by having the drivers restricted. If they were to pay full price then maybe the driver would have been included. It's kind of what Palm did back in the day when they could have spent the extra $.05 for a higher quality speaker but didn't on the Palm III series. I'm pretty sure this is a cost cutting technique used by most electronics manufacturers.
if im paying £500 for a device why not stick $0.5 on the bill! thats a shame
Damn right. I agree with all you guys.
The fact that I cannot get graphics hardware acceleration from a 500 USD+ device is absolutely ridiculous.
I love my Xperia X1, but in hindsight, I would've NEVER supported any product like this had I known it was the case.
Shame on HTC and Sony for swindling their customer from the "rich, multimedia experience" they promise so much.
Bastard Sony/HTC.
well boot android and youll have more qualcomm support..since they own the gpu chip msm7xxx devices uses now, thats what chipset all official android handsets use, even non htc android phones.
ajclai08 said:
well boot android and youll have more qualcomm support..since they own the gpu chip msm7xxx devices uses now, thats what chipset all official android handsets use, even non htc android phones.
Click to expand...
Click to collapse
and thus, there is driver support natively in android for us?
If that's true, what's stopping us from reverse-engineering the android driver and building a WinMo driver?
If we had a starting point, it's feasible--but without a starting point (or white papers), it's impossible.
i completely agree with this we have waited enough for the garbage companies to put out. Ive gone through three HTC phones already with graphics Acceleration turned OFF... WTF!!!
Hi guys! I am not only posting in this forum, I posted the same thing on some other forums too.
THIS MEANS A LOT TO ME AND OTHER PEOPLE WHO LIKE ICS & MIUIv4!
As long as you have a phone with a ARMV6 CPU, please, have a heart, and sign this Groubal now!
This Groubal focus on requesting Qualcomm to release the Ice Cream Sandwich drivers compiled for ARMv6 phones. If the Groubal is signed by large number of people, a complaint would be brought DIRECTLY to the company of Qualcomm.
If you signed this, you already made a difference in the develper world. Let's show Qualcomm what we want!!
SIGN NOW!!
THANKS A ZILLION!!
Up!
Sent From My W8 Using Tapatalk 2!!
http://www.groubal.com/ics-compatible-openmax-il-compiled-for-armv6-from-qualcomm/
What are OpenMAX Drivers?
Those drivers are needed for H/W Acceleration to work fully on our devices
If this petition gets enough signs Qualcomm will release those drivers.
Thanks
Are you sure that they "will" or you meant that they "might" release it with enough signs? It's a huge difference. From what I last read from an official source, they looked into it and they can't. (Which doesn't sound all that plausible for a couple obvious reasons)
Anyway, I signed
Cheers
This is months old, and Qualcomm said they couldn't do it.
Sent from my jelly bean
:'(
too bad, I'll follow this thread so I know if things might change for the best perhaps...
According to Reuters, TI seems to be pulling or re-purposing its OMAP line for use in automotive and other non Mobile applications. I for one was looking forward to the new OMAP5 processor. What are your thoughts?
source: http://www.reuters.com/article/2012/09/25/us-texasinstruments-wireless-idUSBRE88O0WC20120925
shreddintyres said:
According to Reuters, TI seems to be pulling or re-purposing its OMAP line for use in automotive and other non Mobile applications. I for one was looking forward to the new OMAP5 processor. What are your thoughts?
source: http://www.reuters.com/article/2012/09/25/us-texasinstruments-wireless-idUSBRE88O0WC20120925
Click to expand...
Click to collapse
There's a reason no one used their chipset - it lacked an onboard modem. They were reknown amongst developers for open sourcing binaries but they were staring to fall behind.
*Cough* http://forum.xda-developers.com/showthread.php?t=1906607 *Cough*
miketoasty said:
*Cough* http://forum.xda-developers.com/showthread.php?t=1906607 *Cough*
Click to expand...
Click to collapse
some how i managed to overlook that post , my bad. Mods please close/delete this thread.
Ripped from /r/android. Found this to be quite interesting. Apparently the OEMs don't really control if their devices get support for new android versions or not.
http://www.reddit.com/r/Android/comments/1j13xd/how_oems_havent_announced_43_updates_for_the/
https://plus.google.com/116988351660148062102/posts/MHhMo7X1fbF
Shen Ye said:
To all the people complaining about how OEMs haven't announced 4.3 updates for the devices:
• OEMs do not get the Android source code directly from Google.
• The SoC vendors are provided the code from Google, where they make a board support package (BSP) which contains drivers and optimisations etc.
• The BSPs are then passed on to the OEMs, which they use to develop updates for their devices.
OEMs are currently waiting for the silicon vendors to decide which SoCs they will support in making a 4.3 BSP for, because their update support is heavily dependent on this.
For example, Qualcomm recently decided to drop development for a 4.2.2 BSP on their S3 SoCs, which is why HTC had turn around and say they were dropping support for the One S. This also caused Sony to drop update development for the Xperia S, SL, Acro S and ION (all S3 SoCs).
Everyone remembers the Thunderbolt and Sensation (LTE variants) which used the Scorpion MSM8655, which Qualcomm dropped support for, so OEMs had to drop future updates for.
Samsung is an exception, they're their own devices' silicon vendors when it comes to Exynos. But they also had to drop support for their S2 LTE variants which were using the Qualcomm Scorpion SoC.
Sure, it's not the perfect system, but it's how it works in the industry right now.
Click to expand...
Click to collapse
This doesn't sound right if there's people bringing unofficial 4.2.2 updates to said devices. If some guy that that doesn't even have a job with android developing can do it, I think a multimillion dollar OEM can.
Ascertion said:
This doesn't sound right if there's people bringing unofficial 4.2.2 updates to said devices. If some guy that that doesn't even have a job with android developing can do it, I think a multimillion dollar OEM can.
Click to expand...
Click to collapse
True, but none of those 4.2.2 ports have kernel source code, and therefore none of them truly work 100% with no bugs whatsoever. Correct me if I'm wrong, but I believe devs use prior kernel sources and modify them to work with newer android versions (for instance, I had ICS via CM9 on my droid incredible 2, but it never worked 100%).
Yay, time to send hate-mail to Qualcomm.
Sent from my buttered S3
User "iamadogforreal" had an interesting reply:
iamadogforreal said:
This is OEM apologia. In the real world, these OEMs are the customers and boss around the chip makers. Not the other way around, like this blogger is claiming. If OEMs cared about updates then they'd put that in their contracts and pressure the chip makers to do them. Instead, this becomes a convenient excuse (collusion?) for OEMs to stop making those expensive updates, especially when you're 4 months from launching another flagship phone.
Funny how the Nexus line doesn't have this problem. Gee, maybe google just is getting lucky with niceguy SoCs? No, google puts this in their contracts.
Honestly, if you think the SoC guys are telling the world's biggest companies like Sony or Samsung to **** off and to tell your customers to **** off, then you're incredibly naive. Android fans need to keep pressuring these companies to deliver timely updates and to commit to a two year update cycle, at least. Articles like these don't do us any service and only exist to validate that awful status quo.
Click to expand...
Click to collapse