Folks,
I was originally running the stock kernel + TnT Lite (latest) and using quadrant standard was achieving benchmark scores in the 2400's - now after installing pershoot kernel (latest with no modules) and the latest VegaN I am now seeing a score of just over 1700 - is this something I should be particularly concerned about? Are there, for lack of a better term, optimizations that are missing either from this ROM or kernel that are coming in the future?
Stock kernel and vegan tab 5.1 gets me mid 2400's
CPU governor
I believe this is because pershoot uses the ondemand cpu governor instead of performance. Gives better battery, but benchmarks are lower. In daily use you really shouldn't see a difference as the cpu will ramp up when you need it.
that totally makes sense... thanks...
I have a custom compiled ROM from ejhart in #tegratab and I get 2000 in Quadrant but I run Dungeon Defenders flawlessly, so I could give two s**ts about Quadrant
TheJesus said:
I have a custom compiled ROM from ejhart in #tegratab and I get 2000 in Quadrant but I run Dungeon Defenders flawlessly, so I could give two s**ts about Quadrant
Click to expand...
Click to collapse
I too don't necessarily care about break-neck speed results in Quadrant but what I do care about is if lower scores are being caused by bleeding-edge drivers that may cause problems for me down the road vs. power saving techniques being employed by the kernel - hence my original question.
But at the end of the day, if what you need runs smoothly, then its all moot
Rumbleweed said:
Stock kernel and vegan tab 5.1 gets me mid 2400's
Click to expand...
Click to collapse
I am running VEGAn 5.1.1 with the stock kernel also but I am getting scores of 1800-1900 with Quadrant Standard
How can I figure out why I am not seeing a higher level of performance? Thanks.
Jay
Just ran it and got 2559 - Vegan 5.1.1 - Clemsyn Kernel and libsqlite.so and plenty of stuff running.
I'm pretty sure it has always been around 2100+ on any ROM or stock kernel I have tried.
That is strange..
I know that quandrant is not a good benchmark because the results depends on the hardware installed (CPU qualcomm or samsung or TI OMAP), but my question is:
Desire S in quadrant advanced reach a score of 3000 in the CPU section; the xperia arc, with the same CPU reach a score of 5500 in the cpu section. They are both on gingerbread....so why this difference?
Also the desire HD with froyo and the same Qualcomm CPU reach a 5500 score...
I can't uderstaind!
sorry 4 my english
My wife just got her Des S, and Mega jump is lagging
Quadrant score is 1000-1200
how can it happen with an Adreno 205 inside?
it has 1.28.401.1 ROM on it. do you experience it with other ROMs too?
liljom said:
My wife just got her Des S, and Mega jump is lagging
Quadrant score is 1000-1200
how can it happen with an Adreno 205 inside?
it has 1.28.401.1 ROM on it. do you experience it with other ROMs too?
Click to expand...
Click to collapse
i've got no lag at all....
ilbulgaro89 said:
i've got no lag at all....
Click to expand...
Click to collapse
no lag on Mega jump? not even when you pick up a rocket?
liljom said:
no lag on Mega jump? not even when you pick up a rocket?
Click to expand...
Click to collapse
no lag at all in this game
ilbulgaro89 said:
no lag at all in this game
Click to expand...
Click to collapse
and what is your Quadrant score?
liljom said:
and what is your Quadrant score?
Click to expand...
Click to collapse
around 1240
ilbulgaro89 said:
around 1240
Click to expand...
Click to collapse
yeah, that is the max I can get too
the funny thing is, that my Desire HD with a Desire S port performs better; same CPU, GPU and screen resolution; and no lag in megajump :/
liljom said:
yeah, that is the max I can get too
the funny thing is, that my Desire HD with a Desire S port performs better; same CPU, GPU and screen resolution; and no lag in megajump :/
Click to expand...
Click to collapse
yeah, I also do not explain the differences in benchmark scores with this 2 devices
i gave my opinion of why the lower scores in this thread here:
http://forum.xda-developers.com/showthread.php?t=1036668
basically there has been a general trend on most phones that gingerbread performs lower than froyo on benchmark tests and performance. this is certainly been my experience. and since the desire s is running gingerbread, while the desire HD is running froyo (i think, right?), that is your answer right there...
Since the Desire S is the "first" device from HTC dressed in Gingerbread...can we safely say that it wasn't an optimized Gingy version that HTC brought up?
If all the Desire family get their Gingerbread update (Classic, HD, Z) and did perform well in benchmarking, that would leave Desire S...????
RogerPodacter said:
i gave my opinion of why the lower scores in this thread here:
http://forum.xda-developers.com/showthread.php?t=1036668
basically there has been a general trend on most phones that gingerbread performs lower than froyo on benchmark tests and performance. this is certainly been my experience. and since the desire s is running gingerbread, while the desire HD is running froyo (i think, right?), that is your answer right there...
Click to expand...
Click to collapse
you are right
I run Gingerbread on my DHD, and the Quadrant is lower
I thought it's because we don't have the gingerbread sources for DHD, and we don't have the right drivers
but still, DHD has 1500-1600 in Quadrant (compare to 1200) with DS 1.28 port :/ and that's a big difference with the same CPU, GPU
liljom said:
you are right
I run Gingerbread on my DHD, and the Quadrant is lower
I thought it's because we don't have the gingerbread sources for DHD, and we don't have the right drivers
but still, DHD has 1500-1600 in Quadrant (compare to 1200) with DS 1.28 port :/ and that's a big difference with the same CPU, GPU
Click to expand...
Click to collapse
i ask it in this thread for this reason: dhd has an higher results with desire s gingebread porting......a my friend has in test a developer version of incredible s with gingerbread 2.3.3 and his quadrant score is 1450 (higher than desire s...but with same hardware)
ilbulgaro89 said:
i ask it in this thread for this reason: dhd has an higher results with desire s gingebread porting......a my friend has in test a developer version of incredible s with gingerbread 2.3.3 and his quadrant score is 1450 (higher than desire s...but with same hardware)
Click to expand...
Click to collapse
2 things can save us:
-we find the way to root DS
-we get a good update from HTC
DS is relatively still a "baby"
we'll get there eventually
ilbulgaro89 said:
i ask it in this thread for this reason: dhd has an higher results with desire s gingebread porting......a my friend has in test a developer version of incredible s with gingerbread 2.3.3 and his quadrant score is 1450 (higher than desire s...but with same hardware)
Click to expand...
Click to collapse
well maybe your friend's incredible s is overclocked, or he has some other type of customization that you are not aware of. because what i am trying to say is that, stock 2.2 vs stock 2.3, there is a performance and quadrant drop, significantly. stock 2.3 may get around 1250 quadrant, and stock froyo may get around 1650 quadrant. i get 1200 on 2.3, but i can over clock to maximum and get 1400. but somethign still isnt right, because i could over clock on 2.2 and get 1700.
also, people say quadrant scores are meaningless. but this has gotten mis interpreted on the web. quadrant scores are meaningless BETWEEN DIFFERENT PHONE MODELS. but quadrant definitely has meanign within the same device. so if your device runs 1600 on 2.2, then 1200 on 2.3, that is a meaningful piece of data. benchmark shows the ability of your phone to do various tasks like scroll a list, load some html or javascript, or parse a bunch of data. so it annoys me when others (not in this thread) say quadrant is meaningless. what the original statement meant was that lower device A could be optimized very well to be smooth, while higher device B could be better hardware, but optimized poorly to not be smooth, and a benchmark would not tell this story. i agree wtih that. but not within the same device i dont
i also see slower choppy scrolling in 2.3, for example tapatalk threads and lists were smooth in 2.2, but choppy and chunky in 2.3, etc. and the benchmark seems to agree with that.
RogerPodacter said:
well maybe your friend's incredible s is overclocked, or he has some other type of customization that you are not aware of. because what i am trying to say is that, stock 2.2 vs stock 2.3, there is a performance and quadrant drop, significantly. stock 2.3 may get around 1250 quadrant, and stock froyo may get around 1650 quadrant. i get 1200 on 2.3, but i can over clock to maximum and get 1400. but somethign still isnt right, because i could over clock on 2.2 and get 1700.
also, people say quadrant scores are meaningless. but this has gotten mis interpreted on the web. quadrant scores are meaningless BETWEEN DIFFERENT PHONE MODELS. but quadrant definitely has meanign within the same device. so if your device runs 1600 on 2.2, then 1200 on 2.3, that is a meaningful piece of data. benchmark shows the ability of your phone to do various tasks like scroll a list, load some html or javascript, or parse a bunch of data. so it annoys me when others (not in this thread) say quadrant is meaningless. what the original statement meant was that lower device A could be optimized very well to be smooth, while higher device B could be better hardware, but optimized poorly to not be smooth, and a benchmark would not tell this story. i agree wtih that. but not within the same device i dont
i also see slower choppy scrolling in 2.3, for example tapatalk threads and lists were smooth in 2.2, but choppy and chunky in 2.3, etc. and the benchmark seems to agree with that.
Click to expand...
Click to collapse
my friend's incredible s is a stock phone updated to Gingerbread via OTA today...without any type of hack...incredible s is not rooted!
My understanding was that Gingerbread is quite a bit slower than Froyo in Quadrant tests.
In real use though, it seems fine.
I suppose we'll have more idea when 2.3.3 upgrades become available for devices currently running 2.2.x.
njd said:
My understanding was that Gingerbread is quite a bit slower than Froyo in Quadrant tests.
In real use though, it seems fine.
I suppose we'll have more idea when 2.3.3 upgrades become available for devices currently running 2.2.x.
Click to expand...
Click to collapse
Desire HD came out with 2.2, and now it runs 2.3, and yes, the Quadrant went down
njd said:
I suppose we'll have more idea when 2.3.3 upgrades become available for devices currently running 2.2.x.
Click to expand...
Click to collapse
On htc-hero score was equal on 2.2 and 2.3 (both overclocked CM-ROMs)
On my NexuS with 2.3 I have scores about 2500 @ 1GHz (and crazy >3000 @ 1,4GHz)
On the DesireS I only get 1200 @ 1Ghz (equal performance gov or ondemand)
Wired is the totaly different fps: desireS has 50fps at the "planet3D" where Nexus has only 20fps; in the "doom3D" it is vice versa
But in everyday use there is no big difference - I tend to sell the NexuS, but really not sure wich phone I should keep..
Before anyone chastises me: I know that quadrant scores can be manipulated, and shouldn't be relied on to determine if your setup is adequate. I'm talking about my own personal experiences here.
I've only had an android phone for a month or so (my Captivate), so I can't claim to be experienced with it. However, like I did when experimenting with Linux distros (a long, long time ago --- settled on Slackware), I've been playing with all the ROM's available for the Captivate.
Initially, it was to get a feel of the differences in the ROM's themselves and that mostly had to do with user interface. However, I soon figured out that some were faster than others 'native', and then I discovered the Quadrant Scores.
So, I started checking Quadrant scores with each ROM under similar conditions (Wifi on, only a couple of apps running, no overclocking, etc.). This is a really horrible addiction -- I keep thinking that maybe the next ROM, or the next kernel update will make the score better - even though I've noticed that a better Quadrant score doesn't necessarily have much bearing on the performance of the Captivate.
The result: I haven't stuck with a single ROM for two days running.
I seem to have settled down recently, but we'll see.
This is a tongue-in-cheek rant, incidentally...this whole experience has been addictive.
Hi . I made 2 quadrant tests : with eclair2.1 and cm7.1.0 . With eclair2.1 i get 800-900 pcts and with cm7.1.0 1600-1700 . I didn't overclock or other setting made .
Yeah, but that's what I mean - a small change can result in better scores, so where do you stop?
So far, on my Captivate, I've found that I get the best scores with
1) Fasty III - 2500+
2) Dlev 4.2 (Talon) - 2250+ (Dlev 5 doesn't like my Captivate)
3) Biffmod (CM7 based) - 2000+
4) Miui 1.10 + Glitch kernel - 1900+
Here's the issue: I love the UI of Miui - it seems to have the best themes available and everything works smoothly. The stock browser on Fasty III is laggy, for example. The games I play work best on Miui and Fasty.
So, the Quadrant scores don't necessarily have any 'real world' meaning. Most of the lagfixed kernels seem to take care of the I/O rates, improving the scores.
Still, I can't stop trying to find a 'better' setup.
PM'ed reinbeau to close.
Quadrant scores mean nothing, I can enter a single code and get over 3000 on a stock RFS partitioned ROM with no OC/UV (Which is incredible) but it means nothing because the Quadrant app is outdated and wasn't meant for the Hummingbird processor, nor Gingerbread on the phones, or dual core.
seems the higher my score gets the worse my phone runs.
Its all about balance
Thread closed.
I recently switched from hellfire phoenix running faux kernal OC'd to 1.54ghz to Weapon 3.0 R1 and faux kernal OC'd to 1.54ghz.On HP i average a 7800 score using Antutu. When i first put weapon on i did a quick Antutu benchmark to see how it compared to HP i got like 74xx which was a bit lower than my top HP score but OK. Now when ever i try to benchmark i get around 5000-6000.... and i cannot figure out why. Any thoughts? performance is fine, and i don't care to much about benchmarks but i'm just a little curious. I also tried quadrant but i also get a score that's actually lower than the stock rom on the g2x :/
thanks!
They updated the application on the 13th, the scores are different now. You'll score around the same with HP if you reflash it.
cool, thanks