New method to compare up/down load speeds cross platform - HD2 General

I was slum'in over on the EVO forum and there are a bunch of people there asking about Sprint speeds. Many are trying to compare their current platform and provider to make the decision on whether to jump or not. The problem is the cross platform comparisons did not make sense because there were too many variables. We also have many others seeking the same kind of info from the HD2 forum. I made the following suggestion. Maybe we can do the same thing here. It would let people with an HD2 know what types of speeds they could expect from different carriers in different parts of the country. It could also let others on other platforms compare to the HD2's speeds by carrier and location.
My suggestion:
Different browser's get too many variable speeds because of proxy and other problems to make comparisons relative. How about we try this to get closer to Apples to OS's. I believe that Opera Mini is the only browser that is common to the most OS platforms. It at least works on Android, WinMo and Apple now. If all the tests are run on Opera Mini and they are skewed by the proxy than we all will be skewed the same. We can then compare speeds, even though they are not the "real" speeds. The speed itself does not really matter. It is the comparison of speeds that matters.
http://testmyiphone.com/speedtest
is a non-flash mobile speed test that has large enough tests to get results for fast systems. Supposedly the larger the test the more accurate. If everyone who is going to test and compare uses Opera Mini and the"Dual 3MB" or "Dual 6 MB" test at testmyiphone than at least we will have something a bit more relative and scientific to compare. Some will find the 3MB is too small for test to run (fast systems) and others the 6MB will time out (slow systems). Just state which test you used.
It would also be helpful to quote the test speeds as done on your default browser as well so we can see approximately how skewed we are. Just say what browser it is.
OK, save your "who got skewed" jokes and get to testing.
Opera Mini - AT&T - Florida - HD2 T-9193 - WinMo 6.5
6 MB Down - 6.37 Mbps
6 MB Up - 6.08 Mbps
Opera 10 - Default
3 MB Down - 1.32 Mbps
3 MB Up - 0.23 Mbps

Related

only 200mhz?

hi everybody.
i am thinking about to buy a htc wizard (qtek 9100).
i now noticed that the qtek s110 has 416mhz, but the 9100 only 200.
is there a big difference between the two, or is the 200mhz processor as fast as the one with 416?
thx
the Wizard has a differend type of processor than the s110. Its kind of like Intel vs AMD. The mhz doesn't mean it is slower.
I think the wizard is slightly quicker then the intel one but what slows it down is lack of programs using the texas instruments processor extensions and the
non persistant program memory which slows the program loading times quite a lot.
but u get more use out of your batt as it doesnt reserve 30% batt life for ram.
Does the universal lose everything if the battery goes to 0% ? I thought the wizard and the universal were both the same in that they retained their status even on no power ?
knowsleyroader: you are correct. They use persistent memory that will be retained without power. They're slower as a result, but the benefit is what is considered a marginal battery life improvement.
All should read this:
http://blogs.msdn.com/windowsmobile/archive/2005/11/17/494177.aspx
Wizard is generally considered fine for cpu power. I've never used it, but most seem to say it's fine on everything except Skype (which some have found ok). Since you can't get another device with WM5 of the Magician's size (I think), the Wizard is pretty much your only choice right now.
Unless you've got a 700w
V
Pocket Quake runs at a respectable 8.5fps (default settings), and 14.5 (optiomized settings, no sound). I have the Spb Benchmarks (overview below) if you want them.
I find the device slow, but it is not the processor that really slows it down, it is the IO.
On the keeps memory on power-off. I have read reports of the battery needing some fiddling inorder to recharge the battery if you let it run flat.
Spb Benchmark index 232.4 (iPAQ 3650 scored 1000)
CPU index 927.45 (iPAQ 3650 scored 1000)
File system index 94.72 (iPAQ 3650 scored 1000)
Graphics index 2862.38 (iPAQ 3650 scored 1000)
Platform index 273.92 (iPAQ 3650 scored 1000)
intel vs AMD. The mhz doesn't mean it is slower
Click to expand...
Click to collapse
speed is everything my friend
well i had a xda2 mini s had it not even a week and took it back to the o2 shop . and told the guy that the phone is far to slow for what i use it for . so i changed it for a xda2i . and now i am very happy with it never crashes and does not hang up while changing screens like the mini .
intel (r) pxa275
speed 520mhz
128mb ram
thats the speed of my 2i . in my eyes the mini is a phone and just a phone . it cant handle being a pda also . it just dont have the power . and as for the slide out keypad what a joke . my one was starting to get slack in no time . i just could not imagine how this phone would look 6 months down the line of day to day use . be smart and get something that works . ok it looks good but its slooooow
my 0.002 pence worth
musiccube said:
intel vs AMD. The mhz doesn't mean it is slower
Click to expand...
Click to collapse
speed is everything my friend
Click to expand...
Click to collapse
Speed may well be to you, personally i find the functionality more important, but the MHz quote says that processor speed doesn't mean the device operation is slower. it uses a different architecture so the clock speed doesn't need to be as high for the same output (CPS would be a better measure of cpu performance IMO then all processors would be on the same scoresheet regardless of technology or clockspeed)
http://forum.xda-developers.com/posting.php?mode=reply&t
Sorry to drag this off topic a little, but I was reading the XDA-developer encylopedia, which gives claims duel core. Is it? And does duel core in the mobile world mean the same as in the desktop?
Thanks, Mike
Having owned quite a number of Pocket PCs and Windows Mobile devices over the years I have come across this sort of conversation a number of times.
The real point from my experience is that QVGA devices such as the Wizard / Mini S do not need the speed of say a VGA device such as the Exec / HP hx4705. Speed is needed on VGA devices due to the quantity of pixels that need to be updated on the screen. My HP 2210 QVGA device was quicker in a number of ways than my hx4705 and there latter machine was quoted to have more than a 50% speed increase.
I do notice my Wizard slow a little using PocketInformant when I need to filter or search. That to me looks like processor speed. But for that I get a good battery life while using the phone side quite a bit each day (it's a work sim card in there, fully paid for). My Exec however is quicker at data sorting, filtering etc. but relatively slow to update the screen, rotate the screen etc. Exactly the same as my hx4705, also VGA running the last version of Windows Mobile.
I have no reservation in suggesting the Wizard to people wanting to do a bit of everything. I haven't tried playing a film on it yet but I would expect that to be ok as long as the film is encoded to suite. But power users would likely look to either a more powerful solo device or have a second device to compliment it.
I was just lucky that O2 in the UK are offering such rediculous prices for both the Wizard and the Exec that I could get both.
acetuk said:
Wizard / Mini S do not need the speed of say a VGA device such as the
of everything. I haven't tried playing a film on it yet but I would expect that
Click to expand...
Click to collapse
Except the magican (a model down, sports the same size screen) is 400MHz.
That's a very interesting point. But is it the same make of processor? I expect it was running a Xscale processor.
It might well be that HTC decided / realised that for a phone edition model with a small QVGA screen pure raw speed is not really needed. By moving to the TI processor they kept to about the same speed for most real world scenarios (loading programs, looking up contacts, making calls and so on) and then gained elsewhere (received good battery life etc).
All I can really say is that my wizard is faster in certain core areas of these devices (screen redrawing for example) but loses out on pure raw data processing. But as I said at the start of this I don't read the benchmarks. My wizard is a fraction of a second slower than my exec at loading PocketInformant which I can live with.
I think the wizard is positioned correctly after one week of using it. But then I never used the predecessor so I can't compare the two. Having come from one of the fastest non-phone devices on the market though I can't say I'm really noticing the slowness of these devices. Not out in the real world when I am using them.
Now, I just have to sell on my hx4705. And to say that must mean I am happy!
Thanks for the interesting conversation - I'm new here but already feel at home.
well, but I heard alot that the MDAcII with its TI CPU is too slow to open large documents and so is useless for bussiness and that seems to be a problem of the CPU-power!
I've been using a Blue Angel for the last year and actually think that the Wizard is faster for my own use, as a Phone first and PDA second the market that IMHO opinion the Wizard is actually aimed at.
I use SPB to close down apps properly that I use infrequently (word/Excel etc..) and just minimize apps that I use frequently such as Outlook/Phone and my Wizard flies. Even TomTom5 appears to run much faster than on Blue Angel with route replanning completing in the blink of an eye.
Fair enough if you are asking it to run intensive apps a 2i or universal will be faster but those devices are more PDA than Phone whereas the Wizard is the reverse.
For reference mine is an O2 UK supplied XDA Mini S branded device on standard O2 UK rom with all O2 active rubbish removed. Even the battery life beats my SE k750i mobile it has just replaced!
wilesd said:
I've been using a Blue Angel for the last year and actually think that the Wizard is faster for my own use, as a Phone first and PDA second the market that IMHO opinion the Wizard is actually aimed at.
Click to expand...
Click to collapse
Ooh so I'm not alone!
Same here, especially for browsing heavy sites. The BA would take ages formatting the pages, the Wizard also takes time but less. Interface seems faster to me too (once apps loaded - loading time is dependent on the new memory architecture).
Browsing photos with Resco photo explorer is about the same.
The only point on where I can see big difference is video playback. I haven't been able to play a single video without hangups yet, either by using the same ones than I had on the BA or by trying to reencode differently (using TCPMP). A video that would play at 125% on the BA runs maybe at 75-80%. That annoys me because I would like to use it to show videos to people as a demo, which obviously looks less serious if not smooth.
I wonder if that is TCPMP-related or OMAP-related...
BUT, battery life is great!!
Hey guys, i'm considering 'upgrading'(?) from an xda2i to the mini s, and i've noticed you talking about different programs running faster/slower on either device. Was wondering if you could give me a 'rule of thumb' as to which programs would have loss performance in the mini s compared the the 2i?
Thanks
kilrah said:
wilesd said:
I've been using a Blue Angel for the last year and actually think that the Wizard is faster for my own use, as a Phone first and PDA second the market that IMHO opinion the Wizard is actually aimed at.
Click to expand...
Click to collapse
Ooh so I'm not alone!
Same here, especially for browsing heavy sites. The BA would take ages formatting the pages, the Wizard also takes time but less. Interface seems faster to me too (once apps loaded - loading time is dependent on the new memory architecture).
Browsing photos with Resco photo explorer is about the same.
The only point on where I can see big difference is video playback. I haven't been able to play a single video without hangups yet, either by using the same ones than I had on the BA or by trying to reencode differently (using TCPMP). A video that would play at 125% on the BA runs maybe at 75-80%. That annoys me because I would like to use it to show videos to people as a demo, which obviously looks less serious if not smooth.
I wonder if that is TCPMP-related or OMAP-related...
BUT, battery life is great!!
Click to expand...
Click to collapse
Similar experience here - my former PDA is a Dell X50V with a 624Mhz processor and PocketPC 2003SE, at times it's surprisingly sluggish. However, the wizard is more responsive - although I miss the VGA screen in general I prefer the Wizard for browsing.
My only complaint about the processor so far (I haven't tried gaming) is that it's not much good for playing back wmvs. I've not experimented much with it but this is an area where the X50V is very good - it can play highish bitrate WMVs no sweat. The Wizard seems to struggle, even with lower resolution and bitrate wmvs it stutters.
John
About this, how to overclock the wizard ... ??!!
musiccube said:
intel (r) pxa275
speed 520mhz
128mb ram
thats the speed of my 2i .....
Click to expand...
Click to collapse
The Universal has all that too. But, guess what, the overall experience of that "flagship" is slower than the Wizard. Which goes to show, the CPU clock speed has b*gger all to do with how a device performs in the real world.

3g Speed Question

I am running tests on my 8525 in the Winter Park/Orlando Florida area thru slreports/mspeed test - I am getting the following results:
339
459
447
401
428
454
385
443
503
356
476
512
157
337
268
338
450
These speeds are clearly all over the board and do not seem to come close to what others are showing. My question is how do I tell if the slow speeds are due to my phone or due to my network? I would be happy with speeds consistently in the 650-750 range but as you can see, I never really come close.
Thanks for any suggestions on how to resolve this issue and obtain faster speeds.
Although the amount of programs running on your phone could be an issue (if too much CPU time is being used up by them) I think the network is always going to be the limiting factor. Try resetting your phone before each speed test to rule out the phone part.
If it's any consolation, I was always getting 700-900 on my speed tests with my current ROM until a week or so ago. Now I'm getting 300-600, with only occasional bursts over 800. I haven't changed anything on the phone, so logically it must be network-based, which means either more users sucking bandwidth, or less bandwidth available (hopefully temporarily)
Now, if everyone in the Kent area on T-Mobile could please stop using their phones when I want the internet, that'd be great! Thanks.
I'm so sick of Att...
I have seen speeds of 32-930
most of the time 200 or less. BOO for Att's network!
Advertisements had it from 400-700 for the norm, BS. I called them a few times and they say in my market I should be getting the max speed. Yeah right! they should just admit their network is not up to snuff and lower the monthly $ until they can provide steady fast speed.

speed testing on the HD2

Speedtest.net is where I usually test the speed but that does not work on the mobile due to java etc.
My bro has just updated to 'N' Router and 50Meg Broadband with Virginmedia and I wonder how this can be shown on my mobile. Where can I go to test the speed that the HD2 is downloading at so I can compare My Belkin125 MIMO router at 20meg Broadband.
I have done the 'N' Cab update and not convinced that it has really made any difference to the speed of downloads.
TIA
Ricco
Hi, you could try http://www.mobilespeedtest.com/. There's a video of an HD2 test here: http://www.youtube.com/watch?v=XRlx3bsp0ho.
DSLreports has a mobile speed test as well.
I get around 600+ kbps with my HTC HD2 on T-Mobile US.
i got 1026 Kbps in Austria with HSPA
For best results you have to choose http://www.mobilespeedtest.com/test/huge.php
Then it will test with 2 MB and you will enjou your full HDPA-Speed
EDIT:
Now im getting around 1565
Oberoth said:
i got 1026 Kbps in Austria with HSPA
For best results you have to choose http://www.mobilespeedtest.com/test/huge.php
Then it will test with 2 MB and you will enjou your full HDPA-Speed
EDIT:
Now im getting around 1565
Click to expand...
Click to collapse
I get some real bad speeds im in california, will 3 bars and getting 150kbps
Oberoth said:
i got 1026 Kbps in Austria with HSPA
For best results you have to choose http://www.mobilespeedtest.com/test/huge.php
Then it will test with 2 MB and you will enjou your full HDPA-Speed
EDIT:
Now im getting around 1565
Click to expand...
Click to collapse
How the...I get 250-350 on an average...I get full 3G signals and all that...
I live at las vegas.
Update: Oh yay. 101kbps. wow. I'm just above dial-up.
Mobilespeedtest gives values all over the place that reflect more of "where are you living?" than "how fast is my connection?"
I get between 1000 and 2000kbps results on my PC, on a 25Mbps line.
at home via router.. network speed
what download speed do you get?
756kb/sec in manchester, uk with 1 bar of hsdpa on vodafone
Speed 1412.18 kbps
Latency 0.471 seconds
im just a step below Basic DSL which is 1500 kbps
Virgin Media UK, using Basic Home Broadband WIFI
using a 2MB test
Stock 1.66 WWE ROM
2404 Kbps which is about 300 kb/s
4 bars of GPRS
Browser : Opera 10
3625 Kbps which is about 450 kb/s
4 bars of HSDPA
Browser : Opera 10
1673 Kbps which is about 200 kb/s
4 bars of GPRS
Browser : IE
3480 Kbps which is about 430 kb/s
4 bars of HSDPA
Browser : IE
2 MB file tested
Southampton, UK
carrier o2
ROM and Radio in sig
Kris
What browser for speed tests
I've read that speed tests should be done using Internet Explorer. It seems that Opera and Skyfire use their servers and could impact the results. Could someone say if this is truly the case.
dont bother with mobilespeedtest (well unless you live next door to them)
I've run loads of tests from phone and home (v media) and their results are always different and always slow.it reads my ten meg bb as never better than 2 or 3 meg.don't have an alternative though,sorry.
samsamuel said:
dont bother with mobilespeedtest (well unless you live next door to them)
I've run loads of tests from phone and home (v media) and their results are always different and always slow.it reads my ten meg bb as never better than 2 or 3 meg.don't have an alternative though,sorry.
Click to expand...
Click to collapse
Yeah, every speed test seems to suffer in this area.
n11 said:
I've read that speed tests should be done using Internet Explorer. It seems that Opera and Skyfire use their servers and could impact the results. Could someone say if this is truly the case.
Click to expand...
Click to collapse
certainly for sky fire, yes, they are connecting completely on your behalf.opera is a bit different cos it doesn't always user the proxy(think theres a difference between opera mobile and opera mini, plus turbo mode in the new version...)
course, tethering the comp and running speedtest.net (always been bang on for me) from the comp would work.
In the default Opera version on the T-Mo US HD2 I got around 600 - 700kbps.
In IE through mobile speed tests right after the opera test I got 4762kbps
I'll have to tether it later and see what speedtest.net says lol.

[Q] tmo-us slower download speeds

I've been getting slower download speeds on my Android ROMS lately.
I get full H signal all the time.
I'm getting 400 kbps DOWN and 800 kbps UP. I remember I used to get 800+ DOWN too, about 2 months back.
It isn't a big drawback but when I'm tethering, I can feel the slower download speeds. In Windows Mobile, I consistently get between 800 to 1000 kbps.
If I remember correctly, I used to get better speeds using RMNet vs PPP. I believe all the newer ROMs are PPP. I prefer Sense ROMs and I installed the Cotulla NAND today and still get the slower downloads.
Any solutions will be appreciated.
Try the CM 6.1 NAND build...I usually get crappy DL/UL speeds in my area (350-400kbs)but I managed to pull off over 2MB DL's this morning using that one...which I am pretty certain is a RMNET build. I cannot even get those speeds on my other Android devices in the same areas.

Performance vs. CPU Frequency

I had always read that CPU performance increases were logarithmic with respect to frequency, so I decided to do some testing myself and share the results, which do confirm this (I wasn't doubting it, but wanted to see some actual hands-on data, what can I say, I'm a science student). I thought the forum might be interested in seeing it.
System Info:
Ziggy471 BFS Kernel, 4.4.11 build
Das BAMF 1.3.2 ROM
Procedure:
At each frequency I tested, I did 3 SetCPU Long Benchmark tests and 3 Linpack tests. I know these are not necessarily representative of performance, but they do provide a nice numerical comparison. I averaged the 3 results for each frequency and plotted them. The SetCPU Benchmark shows an especially distinctive exponential decay curve, (Lower is better for SetCPU, higher is better on Linpack) but the Linpack one also displays a vaguely logarithmic shape, although not as distinctive as the SetCPU one.
Plots are attached.
Hopefully, this information will at least be found interesting by some of you who were not already aware of the logarithmic curve of performance and maybe even guide decisions on how high of a frequency is worth it. But obviously my results are not the gospel or anything, just the results I obtained for these tests.
I did some similar testing with roughly the same results. After 1.4ghz you typically get diminishing returns. Tested with several roms and several kernels all with same results. currently running perfect storm 1.2 with ziggy's kernel pre-loaded and is very fast...but let's be honest...the phone is plenty fast stock...we really need more battery!!!! lol
thanks for sharing.
Yeah, it seems across all benchmarks the gains really drop out past ~1.4-1.5.
..Coming from the Eris where I was extremely excited to see 5.1 mflops, well, nuff said I guess....
Luv this fone!!
Perfect Storm stock kernel (ziggy's) is set and 1408mhz...my phone benchmarks faster at 1382mhz so that's what i have settled on.
patdroid1 said:
I did some similar testing with roughly the same results. After 1.4ghz you typically get diminishing returns. Tested with several roms and several kernels all with same results. currently running perfect storm 1.2 with ziggy's kernel pre-loaded and is very fast...but let's be honest...the phone is plenty fast stock...we really need more battery!!!! lol
thanks for sharing.
Click to expand...
Click to collapse
Yup. I'm done with the quad wars. I really hope we get better battery life fast.... Also, I noticed the same diminishing returns as well.
Sent from my thunderbolt
Its to the point where speed is a non issue. I'm not running matlab on this phone, I'm playing pokemon. I'm happy with how fast it is, now if i could get my battery up from 12 hours to ~ 18 I would **** bricks.
athorax said:
Its to the point where speed is a non issue. I'm not running matlab on this phone, I'm playing pokemon. I'm happy with how fast it is, now if i could get my battery up from 12 hours to ~ 18 I would **** bricks.
Click to expand...
Click to collapse
Hahaha. You echoed my sentiments precisely.
Sent from my thunderbolt
We might not be seeing diminishing returns.. what if Linpack isn't properly generating a end result because the phone begins to be too fast for it?
We need much longer tests to get more accurate results.
Diversion said:
We might not be seeing diminishing returns.. what if Linpack isn't properly generating a end result because the phone begins to be too fast for it?
We need much longer tests to get more accurate results.
Click to expand...
Click to collapse
do smartbench
It's nice to know that I"m not the only one nerdy enough to do this. I saw similar results.
Nbench will be good for that, it is very accurate and it takes about 10min to complete, its a stability test rolled into one. Its the only way I can accurately gauge my xoom dual core. OOOO, thunderbolt vs xoom in nbench, give me 10 min...
XOOM
Memory 2.86
Integer 3.89
Float 1.17
Thunderbolt 2.0 kernel, no OC
Memory 2.73
Integer 4.07
Float 0.94
surprising results, thought the xoom would be better, must not be fully implementing dual core in 3.0
Based on the graph, looks like all real gains are minimal for the risk above the chips 1.4ghz rating. Good to know that its almost useless to really OC, and just unlock it to the rated frequency.
This is interesting, I'll have to do a little testing myself. Thanks.
I only used results on which Linpack claimed to have a highly accurate result, but it's probably possible that it's not accurate at the higher clock speeds. I kind of doubt it though. The reason I used quicker tests (Linpack & SetCPU) is because I figured it would be a good enough statistic to compare performance of different clock speeds but on the same ROM and kernel, while allowing me to do multiple trials of each clock rate without the "experiment" being overly time-consuming.
The sticking point for me on chalking it up to inaccuracy is the fact that if you fit a logarithmic function to the SetCPU plot, the R² value is .9947, meaning it's almost a perfect fit.
Additionally, I think you start getting more errors and near kernel panics, as I believe they're referred to in *nix systems, at the higher clock rates (1.7, 1.8, etc) where the voltage, in this particular case, cannot be safely increased but isn't sufficient for that rate. Most of what I know about this is just from my experience in undervolting Macs, though. I'm far from an expert.
dacre said:
I only used results on which Linpack claimed to have a highly accurate result, but it's probably possible that it's not accurate at the higher clock speeds. I kind of doubt it though. The reason I used quicker tests (Linpack & SetCPU) is because I figured it would be a good enough statistic to compare performance of different clock speeds but on the same ROM and kernel, while allowing me to do multiple trials of each clock rate without the "experiment" being overly time-consuming.
The sticking point for me on chalking it up to inaccuracy is the fact that if you fit a logarithmic function to the SetCPU plot, the R² value is .9947, meaning it's almost a perfect fit.
Additionally, I think you start getting more errors and near kernel panics, as I believe they're referred to in *nix systems, at the higher clock rates (1.7, 1.8, etc) where the voltage, in this particular case, cannot be safely increased but isn't sufficient for that rate. Most of what I know about this is just from my experience in undervolting Macs, though. I'm far from an expert.
Click to expand...
Click to collapse
Holy crap! Im sure glad there are guys/gals like the ones in this thread to figure this stuff out because I feel like I walked into Trig class all over again. BTW I hated that stuff but in my own dumba$$ speak, this is kinda what I though. What I mean to say is this is a mobile, comparably lowres, linux based system, so how much speed is enough? One would think that at 1.5-1.8 it would be getting "maxed". If not "maxed" then at least enough that 90% of all users would be so thrilled and never see a need for more, that it would be fine. I (like so many others w TB's) would love to see a way to make that battery last just one full day with average usage. Thanks for all the effort you guys are putting in this! Look fwd to great ROMs as the dev work progresses.
Just using the phone, the higher frequencies actually can feel slower. Am I the only one that noticed? I run 1.4 gHz on the LeanKernel 1.9. Did anyone else realize you can bring it to 1.9? It gets really hot, really fast though.
Nope, mine is the same way. My phone is much snappier at ~1.4-1.5GHz than at 1.6GHz+.
I believe this is because the CPU starts derping up when it's at a clock speed higher than it can handle well (which varies from chip to chip); I think eventually when a big enough error happens it results in a kernel panic (where the phone freezes and eventually reboots), but I'm not exactly sure on how it works.
A quick note on the theoretics from someone who studied embedded systems...
Theoretically performance gain should be proportional to CPU frequency (double your frequency, halve your processing time). In practice you often see what you are seeing here, which is the effect of another system on the phone being the performance bottleneck. Without knowing anything about the cpu architecture and exactly how linpack is designed, it's hard to judge where that bottleneck is on the TB, but the point is that performance should be linear with cpu frequency.
Instead of averaging over 3 trials it would probably be better to take the *best* out of, say, 10 trials.
I'm assuming it's somewhat the same as testing runtime of particular algorithms on a computer--you want the best result you get because it actually is the closest to the actual answer. Processors won't "glitch" to being faster than they really are, so a slower result just means that some of the processor was used for something else (another background program perhaps). You will never get all of the processor's "attention" when doing tests like these on your own phone, so go with the result you get when you get *most* of the processor's attention!
Just a thought

Categories

Resources