Related
What on earth is going on? I know where talking about 2 different Moblie operating Systems and devices, but he performance difference explains alot in terms of why the TyTN/Hermes....is underperforming in the Video department, especially with TCPMP.
If this is the case with Video, which can easily be measured for performance, how much more are these Hermes devices lacking in other departments.
Apparently the developers of TCPMP at corecodec are aware of this problem, but it seems to be more related to the Hermes itself.
Below are a few screenshots of 2 different Pocket PCs using the same Samsung 400 Mhz CPU but with completely different results.
Also, I would not necessariliy pay the figures too much mind if the Video performance was at least as good as on the Vario I or the HTC Prophet, but it is not IMHO.
Tell me what you guys think. Maybe I've got it all wrong.
I did have it at all wrong. please read up about the ATI thingy in other post.
I'm dissapointed with the speed too. I'm wondering if it's got anything to do with the 32-bit bus used?
bydandie said:
I'm dissapointed with the speed too. I'm wondering if it's got anything to do with the 32-bit bus used?
Click to expand...
Click to collapse
Hi bydandie.
You may have a point, but then I'm not having any other speed issues. What were the bus speeds on the Universal and Wizard? Do you know? Also;
Samsung (R)2442A [Hermes]
Samsung S3C2440 [HP iPAQ rx3715]
Both run @ 400Mhz
but as we can they don't appear to be identical processors. I'll google around for some info as I'm no expert, but it may still be ROM specific.
What Hermes variant have you got again? and what ROM version?
I was hoping the upgraded 200 extra mhz over the Vario, would be good, apprently it doesn't sound like the video playback is that much better. What is the video playback like ?
This is down to the Hermes not being able to utilise the ATI acceleration technology (can't remember the actual name) properly.
If you try playing a video in TCPMP with the acceleration on, on the Hermes, you will find that the screen shutters like mad!
So you can only use raw video mode for now, which is pretty slow. (But plenty quick enough for 320x240 videos, i.e. most "pocket pr0n")
mackaby007 said:
bydandie said:
I'm dissapointed with the speed too. I'm wondering if it's got anything to do with the 32-bit bus used?
Click to expand...
Click to collapse
Hi bydandie.
You may have a point, but then I'm not having any other speed issues. What were the bus speeds on the Universal and Wizard? Do you know? Also;
Samsung (R)2442A [Hermes]
Samsung S3C2440 [HP iPAQ rx3715]
Both run @ 400Mhz
but as we can they don't appear to be identical processors. I'll google around for some info as I'm no expert, but it may still be ROM specific.
What Hermes variant have you got again? and what ROM version?
Click to expand...
Click to collapse
The Wizard was 16-bit. I'm using the v1605 but with the HTC ROM.
mackaby007 said:
Hi bydandie.
You may have a point, but then I'm not having any other speed issues. What were the bus speeds on the Universal and Wizard? Do you know? Also;
Samsung (R)2442A [Hermes]
Samsung S3C2440 [HP iPAQ rx3715]
Both run @ 400Mhz
but as we can they don't appear to be identical processors. I'll google around for some info as I'm no expert, but it may still be ROM specific.
What Hermes variant have you got again? and what ROM version?
Click to expand...
Click to collapse
Both run at the same CPU speed, but have very different operating systems. According to the screen shot in the first post, the rx3700 tested is running pocket pc 4.21 (2003 second edition IIRC), whereas the Hermes is running Windows Mobile 5. I would guess the rx3700 doesn't have the slowdown related to persistant storage (for example), using some form of flash for memory is going to be slower than RAM.
I have a Eten M600 on WM5 and the samsung 400 proc. the TyTN use also the 400 proc.
When i compare them both on speed then my Eten is much faster in every way. much more responsive. opening and rendering screens.
Altough i prefer my TyTN over the M600, the build-hardware and sound is much better.
I guess it must be the ROM that isn't perfect yet ( and it isn't perfect yet for sure).
I have a Eten M600 on WM5 and the samsung 400 proc. the TyTN use also the 400 proc.
When i compare them both on speed then my Eten is much faster in every way. much more responsive. opening and rendering screens.
Altough i prefer my TyTN over the M600, the build-hardware and sound is much better.
I guess it must be the ROM that isn't perfect yet ( and it isn't perfect yet for sure).
ROM or CPU lacking?
luminus said:
I have a Eten M600 on WM5 and the samsung 400 proc. the TyTN use also the 400 proc.
When i compare them both on speed then my Eten is much faster in every way. much more responsive. opening and rendering screens.
Altough i prefer my TyTN over the M600, the build-hardware and sound is much better.
I guess it must be the ROM that isn't perfect yet ( and it isn't perfect yet for sure).
Click to expand...
Click to collapse
Thanks for your input guys. Luminus your point sounds like there may be hope for us all yet, but having said that, bydandie has switched to the TyTN ROM and I have moved to the I-mate ROM and admittedly we have both both probably experienced some improvements over the ROMs our devices shipped with but the Video playback issue remains, albeit slightly improved too (therefore indicative of being software related).
I spent several hours reading up on the SC32442A Samsung processor and I have to admit, whilst not having enough knowledge to fully understand all the technicalities, it seems that the CPU in the Hermes is one serious 'dude' which is not being used to its Full potential. Apparently its just as quick as a Xscale 520MHz CPU!
I can believe it due to the responsiveness of my device in general (with I-mate ROM), but its just not evident in the Videoplayback department.
I've also tried many differently encoded movie clips like its native MP4 and Divx etc but only negligable differences. I now believe, as posted by bydandie (I think), that its due to the ATI acceleration chip/software not doing what it is supposed and that maybe HTC should look into this as the different ROMs don't seem to make any worthwhile difference in this particular department.
For anyone who's reading this for the first time let me clarify that playback speed is acceptable to watch any well encoded movie (320x240 @ up to 768 kbps) but pausing, forwarding etc then resuming playback is where the problems occurr (for me anyhow). Benchmarking shows that the Wizard (overclocked @240MHz) easily outperforms the Hermes, but not in everything else. Though admittedly I also find the overclocked HTC Prophet @240MHz more responsive all round! Again this seems to indicate that acceleration support is missing across the border.
For the record: I no longer believe TCPMP is in any way at fault. It performs outstandingly on every other device I have ever installed it on, including Samsungs i300 which also had a 400MHz CPU and benchmarked the same videos at well over 300% compared to 120% - 180% were getting on the Hermes!
I give up for now, the problem is way beyond me but hopefully not far off from being found.
Samsung CPU
I'd be interested where you read about this CPU.
It's based on ARMv4 dating back to 2002 and as such is functionally very poor in comparison to the ARMv5 Xscale platforms. This is why it won't allow the current versions of iPlay to operate. It's also unlikely to have the xscale power and performance scaling capabilities as it pre-dated Xscale.
In practise the only thing that has affected me is iplay.
I have just upgraded my 2750 to WM5.
Contrary to what I was told when I purchased the upgrade (many months ago) it is much much faster than the tytn in every respect. It also seems more stable.
For day to day work though - I don't use the TyTN for video - ipod and Archos cover that for me - I find performance very adequate though.
Paul.
Re: Samsung CPU
pgamble said:
I'd be interested where you read about this CPU.
It's based on ARMv4 dating back to 2002 and as such is functionally very poor in comparison to the ARMv5 Xscale platforms. This is why it won't allow the current versions of iPlay to operate. It's also unlikely to have the xscale power and performance scaling capabilities as it pre-dated Xscale.
In practise the only thing that has affected me is iplay.
I have just upgraded my 2750 to WM5.
Contrary to what I was told when I purchased the upgrade (many months ago) it is much much faster than the tytn in every respect. It also seems more stable.
For day to day work though - I don't use the TyTN for video - ipod and Archos cover that for me - I find performance very adequate though.
Paul.
Click to expand...
Click to collapse
Hi again pgamble. I remember you answered an earlier post of mine regarding the CPU. It seems you understand more on these CPUs than I do.
I've provided the linl below to where I read all about the 2442 CPUs (they seem to have variants of it). Hope you understand it better than me. Please let us know what you make of it. Bydandie mentioned that the Hermes uses a 32bit Bus as opposed to the 16 bit Bus used on the first Vario. Do you know if this would negatively or positively affect the speed of data? I would have thought positive, but then I don't fully understand all the technical jargon in the provided link which will probably explain how the technology is utilised.
I hope it doesn't have to share too much of the available data bus/ bandwidth (whatever it is), thus bottlenecking the CPU. I still think at this time its related to the acceleration technology of the ATI Imageon.
http://www.samsung.com/products/sem...ationProcessor/ARM9Series/SC32442/SC32442.htm
http://forum.xda-developers.com/viewtopic.php?t=61370 :wink:
Hermes is a mini-powerhouse!!
Sorry I should have elaborated in that last post which directs you to more accurate information since I started this thread. The Hermes is in fact a mini-powerhouse and even beats the Universal for playback performance under the right conditions.
It's great that the Hermes is powerfull and beats the Universal under the "right conditions"!
But under the "right conditions" DOS beats Windows, it's just a matter on how you compare... :wink:
However, TCPMP still doesn't work as it should on the Hermes. *Standing by for new ROM*
Moskus said:
It's great that the Hermes is powerfull and beats the Universal under the "right conditions"!
But under the "right conditions" DOS beats Windows, it's just a matter on how you compare... :wink:
However, TCPMP still doesn't work as it should on the Hermes. *Standing by for new ROM*
Click to expand...
Click to collapse
The ROM ive settled on (imate) will play back no problems with rawframebuffer set for the video output. Admittedly id prefer to use the ATI acceleration but this temporary measure is still ok for me. Sound playback has improved as it does not stutter to a halt after a pause or skip. (for the record im using a 2pass Xvid encoded avi file with 128kbps MP3 audio) this is evident in both the imate and dopod roms......but not the updated HTC rom.
ATI & The Core Player v1.0
I agree with you both (last 2 posts), but from further investigations, it seems that the problem is not related to the ROM either but onlt to the ATI decoder. I've e-mailed ATI in the hope that they can enlighten or help us all. :lol: Fat chance of that I know, but in the ATI Hanheld Interface, there is a version number v2.30.......maybe that can somehow be updated directly if ATI will offer some support or through a ROM update (though it does seem like firmware) .
Anyway all said and done I'm going to keep looking until I find a way to get that bloody decoder to work permanently as it seriously improves all video formats by at least 100% in benchmarking, which means no dropped frames for High Quality encoded files.
Alternatively keep an eye out for the new TCPMP player (The Core Player v1.0) which will hopefully address all these issues within the coming weeks/months. :wink:
Yer i can imagine a ROM update may include an ATI-written updater within it which will sort out the problems. It DOES sound like something that can be fixed though so hopefully soon we will all have it working correctly. :lol:
So, AKU3.3 ROM did fix the problem with ATI, where you had screen going crazy, but there are decoding artefacts which I belive are down to TCPMP/Coreplayer to fix.
schriss said:
So, AKU3.3 ROM did fix the problem with ATI, where you had screen going crazy, but there are decoding artefacts which I belive are down to TCPMP/Coreplayer to fix.
Click to expand...
Click to collapse
Yer, if you look on the coreplayer forum there is a recommended encoding rate etc... but ive still watched videos no problem even using a different encoding rate.
Hi,
Finally a really in-depth review - really! And like i pointed in the title:
"The X1 comes with a total of 384 megabytes of RAM memory. Only 256 megabytes is visible in the system, but this is because these 256 megabytes is strictly for applications. At boot there’s about 152 megabytes free.
The remaining 128 megabytes of RAM memory is used for both the video graphics and CPU. According to the MSM7200A datasheet, the graphics part of the chipset (presumably the ATI Imageon 2300 or 2700G chip) is capable of delivering up to 4 million 3D triangles per second, and 133 million 3D textured pixels per second fill rate. Furthermore, it supports OpenGL ES - link that up with the large amount of dedicated video memory, and you’ve got an awesome power horse or gaming machine."
Click to expand...
Click to collapse
Best Regards.
Old news bro.
xmoo said:
Old news bro.
Click to expand...
Click to collapse
new news for me though. I know at the system memory it only shows 256MB of RAM, but I didn't know there's 128MB of hidden memory for GPU, etc. In that case, I wonder if it's the same for other HTC device like HTC Touch Pro, or is this an unique architecture for Xperia?
I did observed that the memory after bootup, is much more for Xperia than HTC Touch Pro, maybe that 128MB of hidden memory is making that difference.
zenkinz said:
new news for me though. I know at the system memory it only shows 256MB of RAM, but I didn't know there's 128MB of hidden memory for GPU, etc. In that case, I wonder if it's the same for other HTC device like HTC Touch Pro, or is this an unique architecture for Xperia?
I did observed that the memory after bootup, is much more for Xperia than HTC Touch Pro, maybe that 128MB of hidden memory is making that difference.
Click to expand...
Click to collapse
Yes, it is new to me too as it only states that X1 has 256MB Ram so it has the extra of 128MB memory hidden, no wonder it is faster compared to my previous Touch Pro.
Good Job, SonyEricsson!
xmoo said:
Old news bro.
Click to expand...
Click to collapse
Sorry bro. Been out of the scene lately, and having seen this today was new to me.
That's no proof at all. We already knew that the Qualcomm chip has hardware acceleration capabilities. Even the Diamond and Touch Pro has it but there is a rumor that it isn't utilized because HTC didn't buy the license for it.
Until someone opens up the X1 and spots the exact 128MB memory chip (only the flash memory and the 256MB RAM chip have been spotted as of now) I won't believe anything. Information is rather dubious it needs to be cleared out by facts (and hardware pictures).
Also the source is rather biased too. The "unofficial SE Blog" as they call it, is SE friendly and will write anything will come to their mind to help SE image and sales. I've read the review and while I think it is overall detailed, on some points their objectivity is quite questionable.
Sounds good, but if Sony cant even get the freaking panels to work smoothly with low res screenshots, i doubt any other software will ever use this.
No game developer will limit his games to one phone, so they come up with rather simple games without 3d stuff...its not like with the iphone where millions of people have the same phone with the same hardware/software base...because of that, they have alot of great games that utilize the 3d hardware but i dont see that happening for any WinMo device.
There might an optimzed version of coreplayer or that PS1 Emu, well might.
XavierGr said:
That's no proof at all. We already knew that the Qualcomm chip has hardware acceleration capabilities. Even the Diamond and Touch Pro has it but there is a rumor that it isn't utilized because HTC didn't buy the license for it.
Until someone opens up the X1 and spots the exact 128MB memory chip (only the flash memory and the 256MB RAM chip have been spotted as of now) I won't believe anything. Information is rather dubious it needs to be cleared out by facts (and hardware pictures).
Also the source is rather biased too. The "unofficial SE Blog" as they call it, is SE friendly and will write anything will come to their mind to help SE image and sales. I've read the review and while I think it is overall detailed, on some points their objectivity is quite questionable.
Click to expand...
Click to collapse
damn true, we have nothing but words.
Also, think about it. A whopping 128MB VRAM for WVGA resolution? A desktop PC with 128MB VRAM on it's GPU could play full on 3D games at SXGA resolution easily. Doesn't sound logical to me at all if they put that much VRAM on this tiny device. Way too generous for HTC. 16MB or even 32MB sounds a lot more reasonable and realistic.
^ yep, I've also said that a few times here.....
so in the end I for one, can call SE is a liar here.
I was surfing around wmpoweruser.com and saw this article: http://wmpoweruser.com/?p=2252. It says that Pharos will release this PDA in the Q1 of 2009 and it will rival the HTC Touch HD.
In fact, the specs of the phone are very similar to the HD, but what calls my attention is one really interesting fact: the Pharos phone uses the very same processor as the Touch HD (Qualcomm MSM 7201A 528MHz).
I would like to ask the experts: if the Pharos payed for the video drivers to use hardware acceleration, will it be easy or, at least, possible to import them to Touch HD and we´ll finally have decent video playback?
Thanks!
I think if the drivers for WM existed we would have them by now. I think the GPU is broken and simply doesnt work properly in the Qualcomm CPU or Qualcomm themselves cant get it to work properly, thats why there is only partial support for all its functions.
HTC Touch HD/Diamond/Touch Pro all three of them have video drivers but they are used only with inbuilt players(WMP and video player in gallery). What we need now is to Coreplayer use those DS filters or gain access straight to QTv 'chip' and they said that they finally came to terms with qualcomm about it and now they can start working on it. So one of the upcoming updates(which one I don't know) should finally end this problem but I don't know when it will come.
Here you have link to the topic regarding QTv situation.
The simple test is whether ANY devices on any platform that use the CPU have decent video performance. If not then its clearly a problem with the CPU itself, rather than just the phone manufacturers being cheap.
Qualcomm has other, cheaper CPUs that are just as fast in general terms, but without the GPU, so why fit a CPU with a GPU if your not going to use it?
Sure the HTC Album App can play some files with acceleration, but that is not offering full QTv support and is not providing us with everything the GPU is capable of in theory.
In my view the GPU is broken (or qualcomms implementation is) and therefore writing drivers to fully utilise it is very difficult, which is why there are no fully featured drivers.
Acceleration clearly works - you can play 800*480 high bitrate files in h264 via wmp, but they're a slideshow through other players.
Yes, but its only partial, its not accelerating everything it should. I still say such drivers do not exist, even at qualcomm. Their recent request for windows mobile driver specialists and developers speaks volumes.
It wouldn't be the first time a technology company has launched a broken product on the unsuspecting world, or talked up a product when it really isnt capable of what they claim.
Back to this Traveler 137, it does have impressive specs, looks, gps, and best of all 3G for us T-Mobile USA users! It just won some innovation awards at CES: http://www.pcworld.com/article/1563...ne_offers_navigation_no_network_required.html
I still think that the drivers are here but I think that winmo itself can be blamed for the situation too(using drivers by the system). WinMo should be rewritten for the multimedia so it would bring some universal standards for all video solutions that SoC has to offer. Something like openCL but for mobiles.
The fact that they are looking for programmers and enginners is a good thing and it means that they want to improve the way their products handle wm but it doesn't mean that they don't have people now.
Until somebody gives me concrete proof that the drivers arent there I won't change my opinion. And please don't start talking about GPU problem(which problem I think lies in faulty use or implementation of drivers and that HTC can be blamed cause they take full responsibility of how they configure the device and not SoC maker). I'm only talking here about video acceleration.
Its just that ive seen the same thing happen before, and its usually graphics drivers or memory controllers that suffer. Almost every graphical device we have has some element of broken hardware in it that requires a work around. It may not be the GPU itself, but the interconnect to the CPU, or the way the CPU and GPU interact is just not optimal. Either way something is not right as it is right now.
I very much doubt that there is a driver sitting on a PC in Qualcomm labs that will fix all the issues. I dont believe that no one has been bothered to buy it or that Qualcomm has priced it so high that it wont be used, that makes no sense at all.
HTC are not responsible for the poor driver performance, in fact they have clearly done their own work to work around the issue. What they are responsible for is using the chipset in the first place.
Im waiting to see exactly what Coreplayer ends up like after their improved QTv suport is out. Im willing to bet that it wont make a great deal of difference to anything but H.264 that is already accelerated by the HD native player, but not by Coreplayer currently.
rovex, you might have a point here btw...
I know that the Touch HD runs H264's inloop deblocker in hardware for example. Switching it on or off has no effect on CPU. For pure software decoders it has about 15-25% effect.
Since you can't really apply that specific acceleration to other types of video, it will be interesting to see what other media will be sped up by implementing QTV.
Edit: TL;DR VERSION: 128 of that 512 megs of RAM in ur Epic is leet special-sawce Samsung RAM that pwns other RAMs, and makes ur GPU make moar trianglez in ur gamez.
Forgive me folks, I'm lazy, so I'm posting this more or less verbatim from what I posted in my blog, with a few minor tweaks. It's long and probably boring, so you have my apologies in advance.
Despite my efforts to pull myself away from ARM architecture, Android, and specifically, the mysteries surrounding the Hummingbird processor, I can never really extract myself. One of these days I'll get around to obsessing over something else (hopefully career-related) but until then, I'll let you know what I think I've uncovered as the solution to how Samsung solved the GPU bandwidth issue (which I puzzled over in my original Hummingbird vs. Snapdragon article.)
There have been a few opportunities where I've had to step in and correct people when they post that a Galaxy S phone has only ~320 megs of RAM. It's an error I see made frequently when people use Android system info applications that can only see the 320 megs of volatile memory, despite the fact that the phone does actually contain 512 megs of RAM. We see it happen every time a new Galaxy S phone is leaked, even the Nexus S.
The explanation for this has always been that a certain amount of memory have been "reserved" by Samsung for the Android OS, and that memory is not visible nor available to applications. Despite this, I've never been able to figure out exactly how the system provides the 12.6 GB/sec of memory bandwidth it (theoretically) needs to push out 90 million triangles/sec with the PowerVR SGX540 GPU.
I'm not quite sure how it happened, but in my meanderings across the interwebs, I ran across the following image on ODROID.com of the block diagram of the S5PC110 that they use for their developer board.
(Edit: Image-Shacked... ODROID didn't appreciate me hot-linking their image. Whoops. Use the ODROID.com link above to see the original.)
Careful observation of the POP (Package-On-Package, or "Stacked" circuits) module on the left-hand side shows 384 MB of LPDDR1 and 128MB of OneDRAM, a term I'd noticed on S5PC110 documentation on the list of supported technologies. I'd assumed that it wasn't used. I'd already determined that even though the Hummingbird supports LPDDR2, it only supports it at 400 Mbps transfer rate (which LPDDR1 is capable of) and, with an x32 bus, only allows for 1.6 GBps data bandwidth, a far cry from the 12.6 GBps needed.
So what is this OneDRAM? According to Samsung, "OneDRAM is a fusion memory chip that, can significantly increase the data processing speed between a communications processor and a media processor in mobile devices," and, "...this results resulting in a five-fold increase in the speed of cellular phone and gaming console operations, longer battery life and slimmer handset designs." (Sic.)
Hear hear! 5 times 1.6 GBps still doesn't equal 12.6, but the 12.6 number is a something I arrived at using a lot of assumptions (4.2 GBps bandwidth needed by the PowerVR SGX540 to perform 28 million triangles per second, multiplied times 3 to make ~90 million triangles per second). I'm satisfied that the OneDRAM is that holy grail memory I've been looking for.
Now, how to prove that it actually exists inside my Epic 4G? Remember, the S5PC110 Hummingbird doesn't come with memory built-in; that's something that gets stacked on when the phone is built. The ODROID guys could very well be using a completely different configuration; though that ~320 megs showing up over and over in Android system info apps hardly seems like a coincidence, assuming the difference between 384 and 320 is actually reserved memory for the OS' own system applications. The OneDRAM on the other hand would be reserved primarily for hardware use, such as the GPU as Samsung earlier suggested.
I turned to one of my Android developer acquaintences, noobnl of xda-developers.com. When I showed him what I've run across, (hoping to see if he'd heard of this before, as he has a good handle on Epic hardware) he told me that I had made a good find. He also pasted some kernel code that clearly referenced OneDRAM, proving that the Epic 4G contains this technology.
So there you go folks. The secret is out. The Galaxy S phones are likely able to achieve such amazing graphics performance via a 128 MB Samsung-proprietary high-speed hybrid memory solution. The remaining 384 megs of memory is plain-jane LPDDR1. The total is the promised 512 megs, and honestly, I wouldn't trade the OneDRAM for 128 megs more of LPDDR1 available application memory, but it's interesting how Samsung has kept the OneDRAM solution so quiet. It's likely enjoying the current GPU supremacy of the Galaxy S phones, unfortunately come Cortex-A9, LPDDR2 memory (> 400 Mbps), and dual-channel memory controllers, they will be back on a level playing field. Who could blame them for setting aside Orion and picking up NVIDIA's Tegra 2 SoCs for their next-gen smartphones? It's a fast-moving industry out there, particularly when you don't have Intrinsity any longer as your ace-in-the-hole. Curse you, Apple.
PS - Some of you guys on here know more about this **** than I do. Please feel free to offer suggestions, corrections, and jeers. Though I'm hoping for less of the latter.
Dude what?
Sent from my SGH-T959D using XDA App
Mannnnnnn ....I had a hunch it worked like that
thanks for clearing this up Im glad im not the only one that figured this out!
Are you not compensating for TBDR memory efficiency?
Interesting. Could also complicate porting newer Android versions, at least wih the same efficiency.
I pretend I understood all that you said..and say...Voila..FINALLY!
Yeah I've seen documentation saying the Galaxy S phones have 4Gb of RAM. 1Gb of which is OneDRAM and 3Gb is LPDDR (Idk version). Now, tell us why ODROID and the block diagram can do 1080p and we can't.
EDIT:
thephawx said:
Are you not compensating for TBDR memory efficiency?
Click to expand...
Click to collapse
I just pulled my information from some raw data about the Intel GMA500 (which is an SGX540 disguised as Intel) running at 200 MHz. I would assume the 4.2 GB/s bandwidth needed assumes TBDR is being used.
It's a shaky line of reasoning though. Wish I had some more hard data on the SGX540, and more specifically, the clock rate it runs at in the Epic.
arashed31 said:
Yeah I've seen documentation saying the Galaxy S phones have 4Gb of RAM. 1Gb of which is OneDRAM and 3Gb is LPDDR (Idk version). Now, tell us why ODROID and the block diagram can do 1080p and we can't.
EDIT:
Click to expand...
Click to collapse
Yeah, I could swear I've seen those numbers somewhere but I can't for the life of me find it.
As for the 1080p, that's a good catch; interesting. And, now that I look closer, WiFi doesn't have N support, though I suppose it's possible they used a different WiFi chip.
delete, double post.
Interesting...I guess it makes sense somewhat...though with that much processing why would they then set an FPS limit? is it suppose to be their way of saving energy?
As for 1080p..we have specifications for it..If i were to guess we just don't have the proper drivers...video decoding is done via the ARM Neon..so if anything 1080p would play slow..but it doesn't play at all...so its either locked out intentionally..or the driver is not configured to...I mean it can handle 720p at [email protected] reason why it can't handle 1080p at lower settings...
Wow.. My brains are like oh my gawds!
Interesting tho
Sent from my SPH-D700 using Tapatalk
Another thing. If you look at the Bluetooth certification site you'll see the Samsung Epic is certified for Bluetooth 3.0. Even though the chip in there is a Broadcomm chip that only supports Wireless N and Bluetooth 3.0.
gTen said:
Interesting...I guess it makes sense somewhat...though with that much processing why would they then set an FPS limit? is it suppose to be their way of saving energy?
As for 1080p..we have specifications for it..If i were to guess we just don't have the proper drivers...video decoding is done via the ARM Neon..so if anything 1080p would play slow..but it doesn't play at all...so its either locked out intentionally..or the driver is not configured to...I mean it can handle 720p at [email protected] reason why it can't handle 1080p at lower settings...
Click to expand...
Click to collapse
Video decoding is on the processor? WHAT. THE. HELL. SAMSUNG.
Sounds like you did your homework, i'll buy it. Thanks for sharing. I'll take the fast vid over ram too
tmuka said:
Sounds like you did your homework, i'll buy it. Thanks for sharing. I'll take the fast vid over ram too
Click to expand...
Click to collapse
Homework, yes... but I'd really like to see an ARM engineer affirm it.
gTen said:
Interesting...I guess it makes sense somewhat...though with that much processing why would they then set an FPS limit? is it suppose to be their way of saving energy?
As for 1080p..we have specifications for it..If i were to guess we just don't have the proper drivers...video decoding is done via the ARM Neon..so if anything 1080p would play slow..but it doesn't play at all...so its either locked out intentionally..or the driver is not configured to...I mean it can handle 720p at [email protected] reason why it can't handle 1080p at lower settings...
Click to expand...
Click to collapse
Yeah, I just actually took a look at the S5PC110 User Manual and sure enough, 1080p encoding / decoding at 30 FPS is showing supported on the block diagram:
"1080p 30 fps MFC
Codec H.263/H.264/MPEG4
Decoder MPEG2/VC-1/Divx"
But then, directly below the block diagram, the following is shown:
"Multi Format Codec provides encoding and decoding of MPEG-4/H.263/H.264 up to [email protected] and
decoding of MPEG-2/VC1/Divx video up to [email protected] fps"
That's an odd discrepancy, particularly for an official Samsung processor owner's manual.
Also worth mentioning is that that manual makes no mention of OneDRAM in the memory subsystem breakdown where OneNAND, LPDDR1, LPDDR2, and DDR2 support are outlined, however, it's clearly listed as a supported memory type in the block diagram, and later throughout the manual. Hmm.
Sent from my SPH-D700 using XDA App
After looking through engadgets live blog there was no mention of hardware acceleration being added to android 4.0. With that being said that was my hopes of 4.0 bringing true smoothness that both ios and WP7 have been having all along. Now i know there is 3d animation ( i may be wrong ) but i know that isn't the full acceleration. Question asked does it have it?
Yes, Android 4.0 features 2D Hardware Acceleration (as originally added in Honeycomb), with some improvements.
Additionally, applications can take advantage of the GPU (Photos, Video & Gallery, for example) for on-the-fly transform effects. For example, if you perform edits in Gallery to a photo, it's actually loaded as a texture in OpenGL, and the "effects", or transforms, are applied leveraging the GPU to vastly improve performance.
Likewise, the panoramic "stitching" is GPU accelerated, and video capture (compression) and streaming (transformations, such as silly faces) are GPU accelerated.
It remains to be seen if the GPU is being used for other aspects of the OS, for example, being leveraged by the browser to assist in webpage rendering, etc. However, even as it is right now, it's a massive step up in quality and performance, and should facilitate wonderful UI/UX experiences for ICS even on older devices, like the Evo, Nexus One, Droid X, etc.
Shidell said:
It remains to be seen if the GPU is being used for other aspects of the OS, for example, being leveraged by the browser to assist in webpage rendering, etc. However, even as it is right now, it's a massive step up in quality and performance, and should facilitate wonderful UI/UX experiences for ICS even on older devices, like the Evo, Nexus One, Droid X, etc.
Click to expand...
Click to collapse
Oh really! Are we to believe that somehow the Evo's GPU will be ICS supported for hardware acceleration? Will that require a HTC-specific ICS update, or is it workable for AOSP before HTC codes in their hardware acceleration?
thegregbradley said:
Oh really! Are we to believe that somehow the Evo's GPU will be ICS supported for hardware acceleration? Will that require a HTC-specific ICS update, or is it workable for AOSP before HTC codes in their hardware acceleration?
Click to expand...
Click to collapse
Well, it isn't really a device-specific feature.
Hardware acceleration in the UI is akin to playing a 2D game, like Angry Birds. The system is actually leveraging OpenGL to handle 'displaying', or 'rendering', the game. Likewise, the UI is 'displayed', or 'rendered', the same way.
The way this relates to Android 4.0 is like this: Hardware acceleration was added to Android in Honeycomb (3.0), and in order to take advantage of it, requires a GPU that is capable of supporting OpenGL 2.0 with drivers that are compatible with OpenGL 2.0. If you have a capable GPU with capable drivers, the OS will use the GPU to render the UI, and voila, hardware acceleration.
If any of those components are not available (or perhaps not working correctly), Android defaults to "software acceleration", which is what has always been present in Android for phones. 1.0 all the way through 2.3.7 all use software acceleration. That is, all of the UI elements are rendered by the CPU.
You've probably experienced lag thanks to this--scroll through your contacts list quickly, flip between full home screens, or load up an intensive Live Wallpaper and open your Launcher and try scrolling--you'll probably find slowdown, if not stuttering in places. This is because the CPU is doing the heavy lifting not only for the OS, but also to render the display.
By alleviating this pressure on the CPU, we free up the phone to do work it's better at handling--running the OS. Additionally, because GPUs are actually designed for rendering, they are far, far more efficient at doing so than the CPU. The result? A massive overhead reduction overall on Android on devices, that means improved performance across the board as well as a much more pleasing user experience.
This is fantastic news for devices new and old; but a real treat for those of us with older phones, as the reduced CPU overhead coupled with GPU acceleration should mean noticeable performance improvements, as well as a drastic reduction in stuttering, lag, jittery-ness in the UI, etc.
Best of all, it should be very simple to implement. Most hardware supports OpenGL 2.0 (our Evos do, for example), and most drivers for that hardware also support OpenGL 2.0 (as our Evos do), so it should really be as easy as building Android 4.0 with the appropriate drivers, and then experience the bliss of a hardware-accelerated Android for the first time.
(Note that this explanation doesn't touch on using the GPU for additional benefit, like editting photos, videos, offloading work from the CPU as I touched on above, etc.)
man that was a bunch of good info i needed +1 will be glad when our evos have that much needed acceleration
Dude Shidell thank you so much for that in depth explanation! You covered everything I could have possibly wondered about, haha. A king amongst men, and a god amongst kings, you are.
Thanks alot shidell that really helps me out alot. I just have one question, did anyone here about usb host, on 4.0?
Sent from my PC36100 using XDA App
BmW13294 said:
Thanks alot shidell that really helps me out alot. I just have one question, did anyone here about usb host, on 4.0?
Sent from my PC36100 using XDA App
Click to expand...
Click to collapse
yeah i read that the galaxy nexus was having a usb 2.0? i will check some images to see if its true
Naturally, Ice Cream Sandwich is onboard, with Google finally revealing the version number as 4.0. Other specs include an HD Super AMOLED display (1,280 x 720), a 1.2GHz dual-core processor, 5 megapixel rear camera (with LED flash), a 1.3 megapixel front-facing cam, 1080p video recording and playback, a newfangled panorama mode, a 3.5mm headphone jack and Bluetooth 3.0. You'll also find USB 2.0(right there), 802.11a/b/g/n WiFi, an embedded NFC module, accelerometer, compass, gyro, proximity sensor and even a barometer -- yeah, a barometer. Finishing things out, there's 1GB of RAM, 16 or 32GB of internal storage space and a 1,750mAh battery. info gathered from Engadget listing all the galaxy nexus specs
Happy to share information.
BmW13294 said:
Thanks alot shidell that really helps me out alot. I just have one question, did anyone here about usb host, on 4.0?
Sent from my PC36100 using XDA App
Click to expand...
Click to collapse
Yeah, Android 4.0 builds upon the USB Host support that was integrated in Honeycomb, which means it has native support for a variety of USB devices. Granted, I don't know the depth of devices or support, but it is present.
Nice explanation shi.