See my test reuslt
Play same file at same card.
I update my MDA pro with Jassar rom or dopod 900 asian WWE,but it is slow too.
It cann't play skyforce at normal speed too.
Why?
x50v
TCPMP Version 0.72RC1 Benchmark Results
Average Speed 148.02%
Video Frames 979
Audio Samples 1496401
Amount of Data 4535 KB
Bench. Time 0:22.046
Bench. Frame Rate 44.41
Bench. Sample Rate 67874
Bench. Data Rate 1.7 Mbit/s
Original Time 0:32.633
Original Frame Rate 30.00
Original Sample Rate 44100
Original Data Rate 1.1 Mbit/s
URL \SD Card\[pda][0640x0480][kannagi_op][1100].avi
Size 14728190
Platform PocketPC
OS Version 5.01
OEM Info Dell Axim X50v
Clock speed 624 Mhz
Video output Intel 2700G 480x640 16bits
Video zoom 640x480 -> 480x640
Audio output Wave Output 44100Hz 16Bits 2Ch.
//////////////////////////////////////////////////////////////////////////
dopod 900
TCPMP Version 0.72RC1 Benchmark Results
Average Speed 67.98%
Video Frames 710
Audio Samples 1083527
Amount of Data 3372 KB
Bench. Time 0:34.815
Bench. Frame Rate 20.39
Bench. Sample Rate 31122
Bench. Data Rate 793 kbit/s
Original Time 0:23.666
Original Frame Rate 30.00
Original Sample Rate 44100
Original Data Rate 1.2 Mbit/s
URL \Storage Card\[pda][0640x0480][kannagi_op][1100].avi
Size 14728190
Platform PocketPC
OS Version 5.01
OEM Info PU10
Clock speed 523 Mhz
Video output DirectDraw 480x640 16bits
Video zoom 640x480 -> 480x640
Audio output Wave Output 44100Hz 16Bits 2Ch.
/////////////////////////////////////////////////////////////////////////////////////
dopod 900 with PERFORMANCETWEAKHIGHRESOURCES.cab
TCPMP Version 0.72RC1 Benchmark Results
Average Speed 76.22%
Video Frames 723
Audio Samples 1104685
Amount of Data 3442 KB
Bench. Time 0:31.619
Bench. Frame Rate 22.87
Bench. Sample Rate 34936
Bench. Data Rate 892 kbit/s
Original Time 0:24.099
Original Frame Rate 30.00
Original Sample Rate 44100
Original Data Rate 1.2 Mbit/s
URL \Storage Card\[pda][0640x0480][kannagi_op][1100].avi
Size 14728190
Platform PocketPC
OS Version 5.01
OEM Info PU10
Clock speed 524 Mhz
Video output DirectDraw 640x480 16bits
Video zoom 640x480 -> 640x480
Audio output Wave Output 44100Hz 16Bits 2Ch.
Re: Dopod 900 is very very slow in playback video or play ga
Hi,
Sorry i havent looked through all your results - but in connection with Skyforce.
You need to make sure you've got v1.22 anything below and the game is unplayable on the jasjar/m5000 etc.
I have quite a few game installed and all seem to run ok - the ones that dont are normally not wm5 optimised.
Matt
I recently bought an HTC Trinity also known as P3600.
I installed TCPMP and launch video. Of course I got the problem with Imageon and switched to raw framebuffer.
Image quality was OK (although compression artefacts are more visible on trinity than Wizard) but performance seemed low.
So I ran benchmarks and surprisingly the trinity (cpu= 400Mhz) is much slower than my wizard (cpu= 225 Mhz).
I get 118% for trinity and 152% for wizard for the same movie and the same parameters for TCPMP 0.71.
What can be the explaination???
My HP 4350 with cpu 400 Mhz is well above 250% for this movie...
This makes no sense for me. It's like my wizard is running 200 Mhz. Or maybe Samsung processor is crap but I read somewhere it is good.
below are test results (I stopped them after 30 seconds because it is not necessary to bench the whole movie to see there is a problem)
for trinity:
TCPMP Version 0.72RC1 Benchmark Results
Average Speed 118.48%
Video Frames 943
Audio Samples 1701539
Amount of Data 2883 KB
Bench. Time 0:31.835
Bench. Frame Rate 29.62
Bench. Sample Rate 53447
Bench. Data Rate 742 kbit/s
Original Time 0:37.719
Original Frame Rate 25.00
Original Sample Rate 44100
Original Data Rate 626 kbit/s
URL \Carte de stockage\My Videos\Beur.Blanc.Rouge.FRENCHDRip.XviD-LOST-LuCiFeR_PocketPC.avi
Size 306622884
Platform PocketPC
OS Version 5.01
OEM Info TRIN10000
Clock speed 397 Mhz
Video output Raw FrameBuffer 240x320 16bits Lookup
Video zoom 320x240 -> 240x320
Audio output Wave Output 44100Hz 16Bits 2Ch.
and for wizard
TCPMP Version 0.72RC1 Benchmark Results
Average Speed 152.09%
Video Frames 932
Audio Samples 1687459
Amount of Data 2848 KB
Bench. Time 0:24.511
Bench. Frame Rate 38.02
Bench. Sample Rate 68842
Bench. Data Rate 952 kbit/s
Original Time 0:37.279
Original Frame Rate 25.00
Original Sample Rate 44100
Original Data Rate 626 kbit/s
URL \Storage Card\My Videos\Beur.Blanc.Rouge.FRENCHDRip.XviD-LOST-LuCiFeR_PocketPC.avi
Size 306622884
Platform PocketPC
OS Version 5.01
OEM Info WIZA200
Clock speed 225 Mhz
Video output Raw FrameBuffer 240x320 16bits Lookup
Video zoom 320x240 -> 240x320
Audio output Wave Output 44100Hz 16Bits 2Ch.
meroupow said:
I recently bought an HTC Trinity also known as P3600.
I installed TCPMP and launch video. Of course I got the problem with Imageon and switched to raw framebuffer.
Image quality was OK (although compression artefacts are more visible on trinity than Wizard) but performance seemed low.
So I ran benchmarks and surprisingly the trinity (cpu= 400Mhz) is much slower than my wizard (cpu= 225 Mhz).
.
Click to expand...
Click to collapse
Interesting. I have seen benchmarks for the Omap processor that showed very good graphics performance, even better than much faster processors because of a separate accelerator inside the chip, maybe that is the reason, or maybe TCPMP is just not optimized to use the new Samsung processor to its maximum potential.
from emule
video code------xvid
video quality----1250kb/s
resolution-------672×272
audio code------AC3
audio birates----372kb/s
yes with last coreplayer build .....
yup
Coreplayer will play it.
But the question he asked was...'smoothly' and in my experience it would be no. The bitrate is waaay too high on both the video and the audio, I would say you'd run a benchmark against that file and get it running at about 85%
yes, I mean "smoothly"!
anybody can give a real test?
This was with an 512*384 Xvid movie:
Average speed: 83.66%
Bench. frame rate: 20.059
Bench. sample rate: 40581
Bench. data rate: 1.86 Mbit/s
Orig. frame rate: 23.967
Orig. sample rate: 48000
Orig. data rate: 2.23 Mbit/s
Hell no ! (said with Will Smith voice). If Core player can play the file specified by the OP i'll throw my x1 out the frigging window.
Not sure if this topic has been brought up but I figured I'd shine some light on it as we are on the verge of finally getting our beloved froyo source code. I have read elsewhere that there is an fps cap on these phones. I get the same consistent results with neocore no matter what rom I'm on. Which leads me to believe there is one. Now I know the powervx is dedicated but I figured with all the tweaks with a modded epic verses a stock di18 version there would be atleast SOME difference. Makes me wonder about what it can really be capable of. Everyone drop your pennies in.
Sent from my SPH-D700 using XDA App
I think I remember reading in one of the threads that said it is capped at 50fps. Or something in that neighborhood. It was definitely below the magic 60fps though.
Yup there is one. There was even a custom kernal that released to up it. If course that was for 2.1 and there was no difference at all because as it is I think it's maxed at 56 now. Which we can't notice passed 30 with the naked eye unless directly next to another source with much higher fps.
Sent from my SPH-D700 using Tapatalk
I can notice anything under 50. Being a PC gamer and going for that sweet 60 makes you extremely anal. I'm sure its locked on stock 2.1 and the dk28 based Tom's too eh? Since we have no source there's no kernel out there to unlock the cap correct? Anyone have any guess as to what the powervx540 can REALLY do?
Sent from my SPH-D700 using XDA App
It can do at least 70fps on a specific kernal and over clock...
That makes me happy. I want it.
Sent from my SPH-D700 using XDA App
running_the_dream said:
I think I remember reading in one of the threads that said it is capped at 50fps. Or something in that neighborhood. It was definitely below the magic 60fps though.
Click to expand...
Click to collapse
Probably the biggest factor in the perception of smooth motion given a frame rate is whether there is a consistent delay between frames as well as if the frame rate is constant.
The Epic's LCD is probably 56 hz, hence the frame limiter. In order to maintain a consistent delay between frames, the frame rate would have to be 56 fps or some fps that divides evenly into 56 fps. More specifically, there would have to be ~18 ms between frames, or some multiple. Removing the limiter and reaching higher than 56 fps will not lead to smoother motion, and could make it worse as well as lead to frame tearing (although it might improve other processing in single-threaded games).
Also, given a consistent delay between frames, I would bet that you could not notice the difference between 28 fps and 56 fps; furthermore, I guarantee you that you could not notice the difference between a 56 fps on a 56 hz screen and 60 fps on a 60 hz screen. Therefore, there is no "magic" that makes 60 hz look better; you are probably just used to 60 hz screens where 56 fps looks choppier than the straight frame rate number might imply (due to very inconsistent delays and a dynamic frame rate).
Keep in mind virtually all television broadcasts and movies run at 24 fps in the US, and they look silky smooth. That is because the frames per second count is very precisely controlled, whereas games frequently have very inconsistent frame rates and thus can look choppier at higher frame rates. I haven't thought of this before, but I believe most TVs are 60 hz (or 120 / 240 hz) whereas the video signals are frequently 24 fps. This makes me think that the higher frequency TVs are just making the delays between frames more and more consistent (indeed 120 hz can make the delays perfectly consistent since 24 is evenly divisible into 120), which would further make me think a 24 hz screen should look just as smooth as a 240 hz screen given 24 fps content. This means you are viewing content with inconsistent delays between frames while watching a lot of TV content (albeit ones that correspond to a defined pattern).
That makes me wonder if games ever implement fps "steps" where they either lock the fps down to a multiple of the refresh rate or ensure that the frames are presented according to defined patterns (e.g. showing the same frame twice, then the frame after that once, then the next frame twice, etc.). I would bet locking a game to 28 fps on a 56 hz screen would look much nicer than allowing a 40 fps frame rate. I would also bet such games are perceived as being more efficient / running more smoothly at lower frame rates than other games that do not take such approaches.
EDIT: The Wikipedia article on frame rate says "Many games lock their frame rate at lower but more sustainable levels to give consistently smooth motion." so I was correct in my wondering!
The difference between 28 and 56 FPS is quite noticeable. The idea that it isn't noticeable is actually laughable. 24 fps on films is only acceptable because of motion blur, which is just a cheap trick.
Also, TVs that display higher than 60 fps are just doing interframe interpolation. There's no extra data being displayed, but there are 5-10 intermediate frames generated between actual frames.
The only correct part of your post is that it would be difficult to notice the difference between 56 and 60 fps.
hazridi said:
The difference between 28 and 56 FPS is quite noticeable. The idea that it isn't noticeable is actually laughable. 24 fps on films is only acceptable because of motion blur, which is just a cheap trick.
Also, TVs that display higher than 60 fps are just doing interframe interpolation. There's no extra data being displayed, but there are 5-10 intermediate frames generated between actual frames.
The only correct part of your post is that it would be difficult to notice the difference between 56 and 60 fps.
Click to expand...
Click to collapse
Motion blur is more or less eliminated on modern TVs. They use high quality panels with very quick response times. If there was blur at 24 fps, 60 fps and above would be completely unwatchable. Most gamers rarely experience consistent delays between frames and constant frame rates in games, and so they believe as you do.
I believe most of the 120 hz TVs have the option to turn off those interpolation features and simply repeat frames, and the manufacturers come up with fancy names for their interpolation algorithms.
I stand by all of my theories and claims. I would bet double-blind studies would back me up on this.
thechicgeak said:
Probably the biggest factor in the perception of smooth motion given a frame rate is whether there is a consistent delay between frames as well as if the frame rate is constant.
The Epic's LCD is probably 56 hz, hence the frame limiter. In order to maintain a consistent delay between frames, the frame rate would have to be 56 fps or some fps that divides evenly into 56 fps. More specifically, there would have to be ~18 ms between frames, or some multiple. Removing the limiter and reaching higher than 56 fps will not lead to smoother motion, and could make it worse as well as lead to frame tearing (although it might improve other processing in single-threaded games).
Also, given a consistent delay between frames, I would bet that you could not notice the difference between 28 fps and 56 fps; furthermore, I guarantee you that you could not notice the difference between a 56 fps on a 56 hz screen and 60 fps on a 60 hz screen. Therefore, there is no "magic" that makes 60 hz look better; you are probably just used to 60 hz screens where 56 fps looks choppier than the straight frame rate number might imply (due to very inconsistent delays and a dynamic frame rate).
Keep in mind virtually all television broadcasts and movies run at 24 fps in the US, and they look silky smooth. That is because the frames per second count is very precisely controlled, whereas games frequently have very inconsistent frame rates and thus can look choppier at higher frame rates. I haven't thought of this before, but I believe most TVs are 60 hz (or 120 / 240 hz) whereas the video signals are frequently 24 fps. This makes me think that the higher frequency TVs are just making the delays between frames more and more consistent (indeed 120 hz can make the delays perfectly consistent since 24 is evenly divisible into 120), which would further make me think a 24 hz screen should look just as smooth as a 240 hz screen given 24 fps content. This means you are viewing content with inconsistent delays between frames while watching a lot of TV content (albeit ones that correspond to a defined pattern).
That makes me wonder if games ever implement fps "steps" where they either lock the fps down to either a multiple of the refresh rate or ensure that the frames are presented according to defined patterns (e.g. showing the same frame twice, then the frame after that once, then the next frame twice, etc.). I would bet locking a game to 28 fps on a 56 hz screen would look much nicer than allowing a 40 fps frame rate. I would also bet such games are perceived as being more efficient / running more smoothly at lower frame rates than other games that do not take such approaches.
Click to expand...
Click to collapse
24fps is actually PAL. In the US it's 30 fps (29.97) since we're on NTSC. Sorry to quibble over details.
houkah said:
24fps is actually PAL. In the US it's 30 fps (29.97) since we're on NTSC. Sorry to quibble over details.
Click to expand...
Click to collapse
Ah, I wasn't aware of that. That is actually very relevant because it makes NTSC broadcasts an even multiple of the refresh rate of 60 hz TVs. Although I believe many other sources are still 24 fps.
24fps and 60 fps is quite noticeable in videogames considering each frame is its own image.
Id much rather see my character move his arm in a 60frame animation than half that. It'd be noticeably choppy since there's half the pics.
If you can't see the difference between metroid prime 3 and halo 3 (30fps and 60fps, respectively) then there's something wrong.
Don't compare a movie with video games when it comes to fps because movies don't have to draw the human body animation for animation since god or whatever did that already lol
Sent from my SPH-D700 using XDA App
A_Flying_Fox said:
24fps and 60 fps is quite noticeable in videogames considering each frame is its own image.
Id much rather see my character move his arm in a 60frame animation than half that. It'd be noticeably choppy since there's half the pics.
If you can't see the difference between metroid prime 3 and halo 3 (30fps and 60fps, respectively) then there's something wrong.
Don't compare a movie with video games when it comes to fps because movies don't have to draw the human body animation for animation since god or whatever did that already lol
Sent from my SPH-D700 using XDA App
Click to expand...
Click to collapse
One thing I didn't think about that is also relevant is the amount of motion in the content. Content that varies quite a bit between frames (action movies and FPS games) benefit more from higher frame rates.
That being said, I still believe most people overestimate their ability to detect differences in frame rates. There is certainly nothing "wrong" with you if you can't tell the difference. Only a minority of people truly can.
Honestly my favorite hobby is arguing with people who say that they can notice a difference between 100fps and 60fps on their 60Hz screen.
Going beyond the refresh rate of your screen does nothing for you and actually degrades the image through tearing. Your next best option is half of your screens refresh rate, and then half of that, etc.
back to the topic though, unless the hardware is capped at something absurdly low like 30fps, then it's probably in place for a reason. I don't know what the refresh rate on the amoled screen is, but whatever it is would be a great place to cap the framerate since anything beyond what the screen can show is a waste of power. If the refresh rate is 50fps and they capped the hardware to run at 50fps, then they've done us a favor.
houkah said:
24fps is actually PAL. In the US it's 30 fps (29.97) since we're on NTSC. Sorry to quibble over details.
Click to expand...
Click to collapse
Since we're quibbling over details, maybe you should check your facts.
NTSC DVD = 720 x 480 @ 29.976 FPS
PAL DVD = 720 x 576 @ 25 FPS
As far as 24 FPS (or 24p), that's another story (actually 23.976 FPS and used for theatrical movies).
But seriously, to even try to compare games' to movies' FPS is ridiculous.
They just aren't the same (Dynamic vs Static).
And to get back on topic here, I agree that syncing frame rate with refresh rate is optimal.
There is no benefit in rendering more frames in a second than the screen can display?
With the demise of CRT displays, the days of 120Hz refresh rates are a thing of the past.
And no, my 120Hz TV isn't really displaying @ 120Hz, it's doing some mathamagic to make my brain think that it is.
I also agree with Roisen, they did us a favor if they have enabled "vertical sync" as many gamers know it.
=]
ACS Xtreme Kernel has the FPS cap raised to like 80... 2.1...... now that 2.2 source is live expect big things all over from devs
BopChie said:
ACS Xtreme Kernel has the FPS cap raised to like 80... 2.1...... now that 2.2 source is live expect big things all over from devs
Click to expand...
Click to collapse
I remember someone saying that the readings for GPS were not correct on that kernel. That the cap was not bypassed but the values were masked off to higher numbers.
Sent from my SPH-D700 using XDA App
thechicgeak said:
Probably the biggest factor in the perception of smooth motion given a frame rate is whether there is a consistent delay between frames as well as if the frame rate is constant.
The Epic's LCD is probably 56 hz, hence the frame limiter. In order to maintain a consistent delay between frames, the frame rate would have to be 56 fps or some fps that divides evenly into 56 fps. More specifically, there would have to be ~18 ms between frames, or some multiple. Removing the limiter and reaching higher than 56 fps will not lead to smoother motion, and could make it worse as well as lead to frame tearing (although it might improve other processing in single-threaded games).
Also, given a consistent delay between frames, I would bet that you could not notice the difference between 28 fps and 56 fps; furthermore, I guarantee you that you could not notice the difference between a 56 fps on a 56 hz screen and 60 fps on a 60 hz screen. Therefore, there is no "magic" that makes 60 hz look better; you are probably just used to 60 hz screens where 56 fps looks choppier than the straight frame rate number might imply (due to very inconsistent delays and a dynamic frame rate).
Keep in mind virtually all television broadcasts and movies run at 24 fps in the US, and they look silky smooth. That is because the frames per second count is very precisely controlled, whereas games frequently have very inconsistent frame rates and thus can look choppier at higher frame rates. I haven't thought of this before, but I believe most TVs are 60 hz (or 120 / 240 hz) whereas the video signals are frequently 24 fps. This makes me think that the higher frequency TVs are just making the delays between frames more and more consistent (indeed 120 hz can make the delays perfectly consistent since 24 is evenly divisible into 120), which would further make me think a 24 hz screen should look just as smooth as a 240 hz screen given 24 fps content. This means you are viewing content with inconsistent delays between frames while watching a lot of TV content (albeit ones that correspond to a defined pattern).
That makes me wonder if games ever implement fps "steps" where they either lock the fps down to a multiple of the refresh rate or ensure that the frames are presented according to defined patterns (e.g. showing the same frame twice, then the frame after that once, then the next frame twice, etc.). I would bet locking a game to 28 fps on a 56 hz screen would look much nicer than allowing a 40 fps frame rate. I would also bet such games are perceived as being more efficient / running more smoothly at lower frame rates than other games that do not take such approaches.
EDIT: The Wikipedia article on frame rate says "Many games lock their frame rate at lower but more sustainable levels to give consistently smooth motion." so I was correct in my wondering!
Click to expand...
Click to collapse
Thanks, I like seeing smooth screens but did not know much about the tech behind it...
thechicgeak said:
The Epic's LCD is probably 56 hz, hence the frame limiter. In order to maintain a consistent delay between frames, the frame rate would have to be 56 fps or some fps that divides evenly into 56 fps. More specifically, there would have to be ~18 ms between frames, or some multiple. Removing the limiter and reaching higher than 56 fps will not lead to smoother motion, and could make it worse as well as lead to frame tearing (although it might improve other processing in single-threaded games).
Click to expand...
Click to collapse
Isn't the Epic's refresh rate 68Hz?
http://www.youtube.com/watch?v=5D1cygsc_1M
Skip to 8:50