[Q] EPIC 4G overclocked Better then INFUSE? - Samsung Infuse 4G

So recently my friend brought the infuse 4g and he ran linpack and quadrent and compared it to my epic 4g and noticed epic 4g has alot higher scores. why is this since this phone comes stock with 1.2ghz

what kernel is he running on the epic. I can say this phone smokes the sgs in linpack. not sure what he is talking about there. and did he have 2.3.3 on the epic? that would explain quadrant, if you setup stagefright on 2.3 a certain way on this chip it gets like 1000 more points in quadrant based on its media encoding/decoding alone.
this is a dumb question, is his infuse stock? and as far as linpack could he be comparing it to an evo not an epic?
next time you make a statement like this come armed with numbers. what is his epic quadrant score, what is his infuse score, and what are the linpack scores respectively? I can tell you I have only seen evidence so far that the infuse overclocks better than other hummingbird phones.

Quadrant sucks.
Sent from my Infuse 4G

Dani897 said:
what kernel is he running on the epic. I can say this phone smokes the sgs in linpack. not sure what he is talking about there. and did he have 2.3.3 on the epic? that would explain quadrant, if you setup stagefright on 2.3 a certain way on this chip it gets like 1000 more points in quadrant based on its media encoding/decoding alone.
this is a dumb question, is his infuse stock? and as far as linpack could he be comparing it to an evo not an epic?
next time you make a statement like this come armed with numbers. what is his epic quadrant score, what is his infuse score, and what are the linpack scores respectively? I can tell you I have only seen evidence so far that the infuse overclocks better than other hummingbird phones.
Click to expand...
Click to collapse
well his infuse just a scores above the stock epic 4g and the epic has a score of 2800 running on android 2.2 and can be overclocked to run stable on 1.4ghz to get a score of 3200 and higher on quadrant. and my friend is hesident to root his phone just because he just purchesed it like 3 days ago.

theres a million factors put in to scoring higher benchmarks. if you use a couple of these custom roms, youll notice huge improvements by whats included in those roms. an overclock kernel will get you faster speeds. switching to different governors will up the response in cpu scaling. data2sd will add almost another 1000 points. not to mention changes changes in hardware between the devices and what apps you have running. theres alot of ways to improve the performance of the device and alot of ways to cheat at quadrant. its a good pissing competition yard stick but dont rely on it to tell the full story.

Lol. Benchmarks.
Sent from my SAMSUNG-SGH-I997 using XDA Premium App

benchmarks are good tools if used properly but I'm not sure if the typical xda user can really read a benchmark and learn from it.
those numbers on the epic sound really high. 2800 even on 2.2.1 is hard to acieve on a hummingbird. but his lagfix is probably not ext4. ext2 and nilfs2 file systems will inflate quadrant to no end. atleast with voodoo we have a common yardstick to other android phones even if quadrant stills seems to give the sgs with voodoo a huge advantage over other devices. there are other posible factors but quadrant can be manipulated as it is over sensitive to a few things. I can assure you it is not a hardware issue and the epic has no real advantage. it is only benchmark manipulation and kernel tweeking.
if I told you my captivate gets a 2600 would you think it was faster than the infuse? well mine got that and more. I got about a 2600 on 2.1 @1200mhz with overclocked gpu and oneclick lagfix. ill assure you my infuse is faster with a 1800 on the same benchmark.
infact I have seen more than 3000 on the cappy, that was with gb.
I have seen nexus s and sgs phones with gb and ext2 or nilfs2 file systems get 35-3700 matching the sgs2 and smoking all other dual core phones. does this mean they are actually comparable to thw sgs2? or 50% faster than a dualcore phone?

it helps to also have the advanced version of quadrant to see where the device is falling or exceeding. even still, you can inflate those numbers.

Related

MT4G vs. Nexus S Benchmarks

http://www.anandtech.com/show/4059/nexus-s-and-android-23-review-gingerbread-for-the-holidays/8
I've always tought that the hummingbird would humiliate the snapdragon with some optimisations. But it even performs worse than a sgs (running AOSP/CM)
Sent from my GT-I9000 using XDA App
You have to put into account that 4G uses MySense and loads of bloated ware where the Nexus S use AOSP pure Google experience. Hell load MT4G with AOSP or CM on it, it'll blow the nexus s away.
Sent from my HTC Glacier using XDA App
epsix said:
You have to put into account that 4G uses MySense and loads of bloated ware where the Nexus S use ASOP pure Google experience. Hell load MT4G with ASOP or CM on it, it'll blow the nexus s away.
Sent from my HTC Glacier using XDA App
Click to expand...
Click to collapse
That and it's cheaper to boot. $449 vs $529
I think if mytouch 4G would run CM or some vanilla Froyo w/ OC, It blow away Nexus
epsix said:
You have to put into account that 4G uses MySense and loads of bloated ware where the Nexus S use ASOP pure Google experience. Hell load MT4G with ASOP or CM on it, it'll blow the nexus s away.
Sent from my HTC Glacier using XDA App
Click to expand...
Click to collapse
The MT4G already blows the Nexus S away. The benchmarks are a fairly good indicator of that.
Yeah, make sure you are reading those benchmarks right, the MT4G kills it.
I think the mt4g is faster no need for benchmarks I notice it on everyday use considering I've had the vibrant and the ns is just a rebadged galaxy s.
Sent from my HTC Glacier using XDA App
Heck, I just think it's cool to own a device that performs so well compared to other high end devices. After owning a G1 for over a year, having the MT4G is a treat!
These benchmarks make my mouth water... Now ****ing xmas needs to be here so I can have my mt4g (that or the wife needs to develop a heart, I should pack up her G2 & tell her she can use our old MDA until Xmas)
The bench marks only show how incomprable the two processors are. And how unreliable benchmarks are. Dont trust neocore. The phones are so fast they cap out the built in fps limits of the phones. Interesting how the losing phone in neocore get 10 fps more running quake. Thats almost 25% more. Also interesting that quadrant cpu is right on par but linpack is still low. I have no idea why qualcom chips do so well in linpack. They obviously have a distinct advantage there. Yet hummingbirds some how do as well on froyo roms in quadrant.
I/0 scores vary wildly with galaxy s phones. Rfs always give a very low score. But there have been "lag fixes" that blow the score out of proportion well beyond anything else in quadrant. CM although buggy gets huge scores all over cept for linpack. A CM dev said the low I/O was something samsung did. It may be that the nexus s has the same problem.
Im not putting the mt4g down. Not at all. Just the benchmarks. I think the galaxy phones still have the best gpu. But if I could get a mt4g unlocked or a nexus s it would be a hard decision. ( im on ATT)
Im partial to samsung after doing repairs to my captivate and an evo. I felt that the htc was built like a toy. But as far as user experience hatch wins every time.
Sent from my SAMSUNG-SGH-I897 using Tapatalk
Dani897 said:
The bench marks only show how incomprable the two processors are. And how unreliable benchmarks are. Dont trust neocore. The phones are so fast they cap out the built in fps limits of the phones. Interesting how the losing phone in neocore get 10 fps more running quake. Thats almost 25% more. Also interesting that quadrant cpu is right on par but linpack is still low. I have no idea why qualcom chips do so well in linpack. They obviously have a distinct advantage there. Yet hummingbirds some how do as well on froyo roms in quadrant.
I/0 scores vary wildly with galaxy s phones. Rfs always give a very low score. But there have been "lag fixes" that blow the score out of proportion well beyond anything else in quadrant. CM although buggy gets huge scores all over cept for linpack. A CM dev said the low I/O was something samsung did. It may be that the nexus s has the same problem.
Im not putting the mt4g down. Not at all. Just the benchmarks. I think the galaxy phones still have the best gpu. But if I could get a mt4g unlocked or a nexus s it would be a hard decision. ( im on ATT)
Im partial to samsung after doing repairs to my captivate and an evo. I felt that the htc was built like a toy. But as far as user experience hatch wins every time.
So you're saying those guys over at AnandTech one of the most trusted reliable source don't know what they're doing,LOL. Even while they're doing a reviews about Nexus S they were amaze at MT4G. Talk about HTC Phone bieng Like a toy,look at the Vibrant for 500.00 buck it's a joke with that kind of money i expect not to be lagging and GPS works out of the box,and don't tell me there's a fix online it's easy to fix.Like one of my MT4G user said it real well, Nexus S is just a rebadged of the
plastic Vibrant Galaxy S with gingerbread 2.3 LMAO!!!!!!!!!
Click to expand...
Click to collapse
blah blah blah double post...
I want to see those mt4g benchmarks with gingerbread. I do agree that benchmarks should be taken with a grain of salt, but it's always nice to come out on top
Yes and they should just be glad they did not have a OC kernel with a trimmed ROM or that nexus S would have been smacked and knocked back to kindergarten.
I have been confused at why the MT4G has not challenged or even been compared to the Nexus S??? All this waiting and hype about the Nexus S and finally it comes out and (IMHO) is a disappointment. Now reading what everyone is *****ing about, No HSPA+, expandable memory, ram and not made by HTC.... Well MT4G has all those!!! Minus the SAMOLED, I think the MT4G is THE better phone. Now with some numbers to look at...Its looking sweeet! Best part is, "The 45nm Snapdragons with Adreno 205 GPUs have proven completely competent performers (and I'm not even finished testing the myTouch 4G yet)"!!!!!! Can someone please put these two in the octagon!!
This sums it all up
I got the MT4G but also want the Nexus S just because I know it will have more developer support than the MT4G , I dont care about benchmarks just user experience, lets see the people reactions when they get it
Sent from my HTC Glacier using XDA App
After exchanging my original MT4G for one with a "superior" display, this phone is most certainly a force to be reckoned with. Yes, the SAMOLED display is great, but highly overrated. I remember doing a side-by-side comparison of my old Vibrant with a G2, only to find that the G2's display more than held its own next to the Vibrant. Even with its washed out display, my original MT4G outclassed the Vibrant in every aspect - the new one, with it's better display, is not even in the same league.
Since it's been established that the Nexus S is basically an aesthetically revised Galaxy S, it's safe to say that the MT4G vs. Nexus S comparisons will yield the same results as MT4G vs. Vibrant. Say what you want about benchmarks, but numbers don't usually lie (unless, of course, you're Bernie Madoff).
That's my reason for getting the nexus s too. Its the rom support and user experience. I've had this phone since almost the day it came out. Its got every feature you could want, but though it looks and feels expensive, htc has major qc issues. I've yet to see one without a blown out white screen either. If they had blessed the phone with a better display, changed the plastic below the screen a little, and made it with asop, the mt4g would be the best phone ever made.
Battery life is excellent, it's fast, and has all the latest hard ware. Just upsets me that with all the specs this thing has, they give it such a poor display. I mean you look at that every time you pick up the phone, and espresso sense themes expose that even more with all the greys and off blacks.
Sent from my HTC Glacier using XDA App

Quadrant scores

I'm seeing consistently over 1600 scores on standard Quadrant on my stock, unrooted Inspire running 2.3.3.

			
				
Lol
..... ten figures
sent from my Inspire 4G running ClearDroid 1.1.51
Hiyabuz said:
Lol
..... ten figures
sent from my Inspire 4G running ClearDroid 1.1.51
Click to expand...
Click to collapse
What, that's perfectly legit synthetic benchmark score I posted.
Everyone knows quadrant can't be made to show better than the phone should be capable of.
Saved and Texan by the grace of God, Redneck by choice.
Just thought it was funny is all. Highest I've seen on mine is around 1700 but I'm not overclocking or anything
sent from my Inspire 4G running ClearDroid 1.1.51

			
				
I usually hit 3k no big deal...
There is no perfect way to calculate the power of a phone and it is completely subjective. If you love the phone then that is a good enough reason to have it.
I know there are cooler phones with Tegra 2's, but why do I need that in my phone? I mainly use it for texting and web access and/or tethering to my Transformer.
OC @ 2.0ghz Performance governor.
Ok guys.we all know quadrent scores mean bupkiss. So no need for these threads LOL
Thread xlosed

[Q] HTC Vivid poor quadrant?

As I already know Quadrant isn't the best benchmark, but I noticed that the Evo 3D which seems to be about the same spec wise as the Vivid gets better scores and especially in FPS almost double at times, almost as if the Vivid is capped
Is this because Quadrant isn't optimized for the Vivid? I'm guess that the answer but would like to know just for sake of cursorily.
Thanks!
Quadrant isn't optimized for any dual core devices. Its an extremely inaccurate benchmark. Looking at specs, they should be pretty close if using a good benchmark test, unless there's a big difference in the gpu.
Sent from my PG86100 using XDA Premium App
Seems like quadrant hasn't been changed in a long while considering the phones on the result list.
I'm pretty sure they share the same GPU, just seems like the phone is getting a bad wrap for that reason.

Quadrant Scores

Question of the day. Why do quadrant scores suck on this device? Gs2 I'm getting above 5300s and this dude has a better processor. It should be killing the Gs2 and I'm only getting 3200s. Uncalled for I say! Maybe someone with better knowledge can explain?
Sent from my SGH-I727 using Tapatalk
Our processor is not better. The gs2 has the best there is right.now. and our quadrant does not suck ove gotten over 4000 .
---------- Post added at 11:46 PM ---------- Previous post was at 11:44 PM ----------
Ours is very good and even though its clocked at 1.5 and the gs2 only at 1.2 the exynos is the best on the market period
+1 on that. Exynos is better. The OC'd gs2 is getting over 9000 on some benches like cfbench. But we have LTE so I'll take a slower bench and faster dl/ul speeds
Interesting. I didn't know the chipsets in the two were that much different.
Sent from my SGH-I727 using Tapatalk
Not to mention quadrent scores mean nothing at all as they can be boosted but make the phone lag. Only real test is real world use. All benchmark scores can be fooled into giving a high score without improving real performance
+1 as i have stated previously. Quadrant scores are like puting an elevator in an outhouse.
silver03wrx said:
+1 as i have stated previously. Quadrant scores are like puting an elevator in an outhouse.
Click to expand...
Click to collapse
And I could use both right now......
Sent from my SAMSUNG-SGH-I727 using xda premium
cdshepherd said:
And I could use both right now......
Sent from my SAMSUNG-SGH-I727 using xda premium
Click to expand...
Click to collapse
I'm a little concerned by that need of yours...
I agree with what was already stated.
While synthetic benchmarks are handy to track relative performance increase/decrease of a device in certain situations, they shouldn't be a deciding factor in anything. They are one source of information that should contribute to an overall decision about device performance. They should be weighted heavily with real-world use and experience.
The Adreno 220 is a better core than the Mali 400, but the Scorpion is a beefed up Cortex A8, so it's not as efficient as the Cortex A9 in the Exynos.
It's debatable as to which is more powerful in real world situations. They're pretty comparable. Skyrocket might have a slight edge in games, vanilla SGS2 has the edge in benchmarks. They're pretty close all around, though.
Frogacuda said:
The Adreno 220 is a better core than the Mali 400, but the Scorpion is a beefed up Cortex A8, so it's not as efficient as the Cortex A9 in the Exynos.
It's debatable as to which is more powerful in real world situations. They're pretty comparable. Skyrocket might have a slight edge in games, vanilla SGS2 has the edge in benchmarks. They're pretty close all around, though.
Click to expand...
Click to collapse
+1
Quadrant scores are quite arbitrary and not reflective of the real world performance. For example, an SQL hack can get your score up by a few hundred points without a real world performance boost.
By the way, I never managed to get more than mid 3000 with the stock build on my SGS2.
enigma00 said:
+1
Quadrant scores are quite arbitrary and not reflective of the real world performance. For example, an SQL hack can get your score up by a few hundred points without a real world performance boost.
By the way, I never managed to get more than mid 3000 with the stock build on my SGS2.
Click to expand...
Click to collapse
Came here to say this. Also in real world usage I'll take a bigger screen and LTE over a couple fps on a graphically intensive game.
Sent from my SAMSUNG-SGH-I727 using XDA App
My best scores:
usually im around 2800-3100 but sometimes it gets crazy fast and hit: (btw the slide to unlock thing is from a security app that i was playing with, i assure it its a gs2 skyrocket)

QUADRANT gone MAD

No one mentioned this here?
Since it's not a Question, I installed the latest Quadrant "cause i blamed it on a ROM {RAVEN}"
Fine, this time it wasn't his fault...
Tested it on SGS4G
SGS 2
Evo4G
LG G2x
ALL show the same, Graphics suuukkkk
And it takes longer than the USPS to deliver a lost package just to get the CPU.
You're welcome
What? none of you noticed this?
Confirmed on a friends brand new SGS4G
CSOS said:
No one mentioned this here?
Since it's not a Question, I installed the latest Quadrant "cause i blamed it on a ROM {RAVEN}"
Fine, this time it wasn't his fault...
Tested it on SGS4G
SGS 2
Evo4G
LG G2x
ALL show the same, Graphics suuukkkk
And it takes longer than the USPS to deliver a lost package just to get the CPU.
You're welcome
Click to expand...
Click to collapse
Which latest quadrant (not to be a douche, just wondering which one is not working well)? Is it just the benchmark test, or is your ROM sucking as well? Remember, quadrant scores don't mean a lot, it's the end user experience that counts. The only reason we posts quad scores is to give peeps a reference of how silky smooth something is. If you are not having the same experiences as others on a ROM, then you should consider Gremlins Removal.
I get 3008 with the new one so..hmm
Sent From My Rockin' Galaxy S 4G
Quadrant 2.0 give lower scores I get 2000
Sent from my SGH-T959V using xda premium
And even at 2000 that is well above the original score of 900.
thomas.raines said:
Which latest quadrant (not to be a douche, just wondering which one is not working well)? Is it just the benchmark test, or is your ROM sucking as well? Remember, quadrant scores don't mean a lot, it's the end user experience that counts. The only reason we posts quad scores is to give peeps a reference of how silky smooth something is. If you are not having the same experiences as others on a ROM, then you should consider Gremlins Removal.
Click to expand...
Click to collapse
I think if I interpreted his post correctly he is complaining about the look and how long it takes not the actual test.
I could be wrong and probably am but that is how I understood him.
CSOS I noticed it but figured it to be a "feature" of the free version.
Yes eollie, that is what I meant. I know i write differently.
Thanks.
even though the scores are different with the new quad if you notice the phone is still higher than the galaxy nexus which has ICS 4.0
dsexton702 said:
even though the scores are different with the new quad if you notice the phone is still higher than the galaxy nexus which has ICS 4.0
Click to expand...
Click to collapse
Weird thing is the scores are WAY different like Im getting 2100 now vs. 2990 before.
Tha antutu score is still the dame but quad is just stupid now.
What you guys don't understand is that all the scoring systems are different. Say I want to make part of your score based off of your SD card read speed, and then multiply that by 10. If you have a class 10 card you might get 16 megs / s read speed, multiply that by 10 and we get a score of 160. I release version 2 of my benchmark and decide that now the sd card read speed will be multiplied by .5, so instead of giving you 160 more points for your sd read speed you only get 8 points.
All the benchmarks are artificial numbers taking measurements that the benchmark author decided. They almost never reflect real usage performance of your phones. You're better off measuring every day tasks like copying your music collection to your phone, phone bootup time, how much time it takes to open your browser, the fps you get in your favorite games, etc. Things that you ACTUALLY do on your phones. Based on this you can have your own measurements to see how your phone performs with different roms.

Categories

Resources