Question Storage benchmarking: sequential write - Google Pixel 7 Pro

Read notebookcheck.net's P7P review and noticed they used AndroBench to test the storage speed so I thought what the hell, I'm gonna do the same. Didn't' fiddle with any settings and just ran the test as is. The sequential write speed is noticeably lower than what the fine people at notebookcheck got (875 MB/s). Also ran PCMark's storage test and the internal & external sequential write speeds were below the expected level.
I am a total novice at this and chances are I'm both improperly configuring/running the tests and interpreting the results. Before running the benchmark I was trying to remedy the so-called scrolling issue so I installed the december factory image using the Android Flash Tool; managed to do that without a hiccup, but now I'm scared the flashing process was somehow detrimental to the P7P's storage health or speed.
I'm guessing the storage drives on phones are resilient and one could, theoretically, flash an image dozens of times in a row without any ill effect , but, as I mentioned, I don't know which way is up regarding this stuff so I'm seeking out the advice of more experienced and knowledgeable users.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}

That's pretty fast. The real question is in actual usage how much is scoped storage slowing it down and how much increased battery usage is it causing?
Write tests put wear on the flash cells while serving no tangible purpose if only done out of curiosity.

Well, been treating my device as if I had it for years, flashing away on it more than 2-3 dozen times now, but my numbers aren't far from yours. I honestly wouldn't worry much about it.

Interesting. I couldn't resist These are the results from my N10+ N975U running on Pie. In spite of slower memory (12gb) and SoC it did very well on sequential write times. I use to think my Raptor x 4 Raid 0 array was fast...

I'm on the original October Firmware and this is what I get, I've been on all of the different versions (November and December) but I get the best battery life on October. Perhaps they changed some write settings in the November or December firmware.

wiggumpd1 said:
I'm on the original October Firmware and this is what I get, I've been on all of the different versions (November and December) but I get the best battery life on October. Perhaps they changed some write settings in the November or December firmware.
Click to expand...
Click to collapse
Interesting... I too flashed the October version this morning and got pretty much the same benchmark results.
Could it perhaps have anything to do with the P7P storage options? Mine is 128GB.

Chuckles_101 said:
Interesting... I too flashed the October version this morning and got pretty much the same benchmark results.
Could it perhaps have anything to do with the P7P storage options? Mine is 128GB.
Click to expand...
Click to collapse
False reading on the sequential write for sure for non raided single drive, I think. Makes me wonder if my benchmark results are valid... maybe... as internal transfers are fast Copy/paste a 1-10gb file and see how long it takes to write it from internal to internal.
Real world speed is all that counts...

Mine is even worse. Running the beta.
Look at the first one I ran too. Eek.
Edit: I also ran another app, CPDT Benchmark, to see if it was just the app. It's definitely not.

Just to clarify my last post: I got the same results on both October and December versions.
As blackhawk suggested, I copy/pasted a 5,3 GB video file a few times to test internal to internal. On average it took around 26 sec, which would give me roughly 205 MB/s.

EtherealRemnant said:
Mine is even worse. Running the beta.
Look at the first one I ran too. Eek.
View attachment 5794365View attachment 5794363
Edit: I also ran another app, CPDT Benchmark, to see if it was just the app. It's definitely not.
View attachment 5794369
Click to expand...
Click to collapse
Random write is faster than random read? This some kind of alternative universe? Or is there something about flash memory that eludes me?

Chuckles_101 said:
Just to clarify my last post: I got the same results on both October and December versions.
As blackhawk suggested, I copy/pasted a 5,3 GB video file a few times to test internal to internal. On average it took around 26 sec, which would give me roughly 205 MB/s.
Click to expand...
Click to collapse
About 24 seconds to copy a 5.2gb jpeg folder.
Couldn't find a large enough single file.
Not sure I trust that app...

blackhawk said:
Random write is faster than random read? This some kind of alternative universe? Or is there something about flash memory that eludes me?
Click to expand...
Click to collapse
Looks like a caching issue.
Why are 4k reads in hdd/ssd benchmarks slower than writes? - HDD FAQs
Question: There are several benchmark tools available to test the speed of a pc drive. Here is an benchmark example of a SATA SSD: Sequential Read : 718.498 MB/s Sequential Write : 777.414 MB/s Random Read 512KB : 160.541 MB/s Random Write 512KB : 838.930 MB/s Random Read 4KB (QD=1) : 26.985...
hddfaqs.com
NVMe ssd: Why is 4k writing faster than reading?
I have a Samsung 960 Pro 512 GB SSD on NVMe with PCIe Gen.3x4 running. I use the Samsung NVMe Driver 2.0.0.1607. The SSD is running fine. However, I don't understand why the writing of 4k is faster...
superuser.com

EtherealRemnant said:
Looks like a caching issue.
Why are 4k reads in hdd/ssd benchmarks slower than writes? - HDD FAQs
Question: There are several benchmark tools available to test the speed of a pc drive. Here is an benchmark example of a SATA SSD: Sequential Read : 718.498 MB/s Sequential Write : 777.414 MB/s Random Read 512KB : 160.541 MB/s Random Write 512KB : 838.930 MB/s Random Read 4KB (QD=1) : 26.985...
hddfaqs.com
NVMe ssd: Why is 4k writing faster than reading?
I have a Samsung 960 Pro 512 GB SSD on NVMe with PCIe Gen.3x4 running. I use the Samsung NVMe Driver 2.0.0.1607. The SSD is running fine. However, I don't understand why the writing of 4k is faster...
superuser.com
Click to expand...
Click to collapse
My drive doesn't reflect that. More in line to what to I would expect. I generally just look at how fast large file transfers go. On internal memory they go fairly fast.

Related

Make your GALLERIES & MUSIC load FASTER!

NEW PROGRAM MADE SPECIFICALLY FOR DEFRAGGING FLASH MEMORY:
Yep, I know, they say defragging flash memory is unnecessary and wears out your SDCard. But then again, a car that can go 160 mph is unnecessary too, but that doesn't mean you don't want it.
My favorite (and free) defragmentation program is called:
MyDefrag << Click to D/L.
You can get it online and it goes on your PC (not your phone):
It has a NEW option that is made specifically for defragging flash memory (i.e., your SDCard). I figured, what the hell and gave it a run.
NOTE: THIS PROCESS CAN GET HUNG UP ON YOU (IF IT DOES, DO THIS)
It took 3 tries to get mine to go all the way through the first time. Just exit out of MyDefrag and when restarting chose the "Kill Current Session" option and it will pick up where it got stuck before.
This probably works faster if you pull your card and put it into a usb card reader. Of course, back up your SDCard on your PC before trying this.
THE RESULTS:
1) Even large galleries populate VERY fast.
2) Music and video libraries load instantly.
3) When adding apps/widgets to homescreens, options appear MUCH faster.
3) NO improvement in read/write speeds (but I didn't expect that anyway because the read/write test is just reading and writing 200 mb of data to empty space anyway).
CONCLUSION:
This takes a while so only do it when you have a spare 1/2 hour or more that you don't need a phone. As far as wearing out your SDCard, c'mon, we are flash-a-holics and are beating our cards to death daily anyway. However, I would recommend doing this only monthly or so.
* P.S. - This is also an excellent defragmentation tool for your PC.
** UPDATE: We have been testing this out over at the MikFroyo thread and members have been getting excellent results.
IF THIS THREAD HELPED MAKE YOUR PHONE FASTER, GIVE IT A 5-STAR RATING ON TOP SO OTHERS WILL SEE IT TOO ​
Hell, I'll try anything once. Thanks for this.
trying now..i followed it on MikeFroyo post.... .. like he said above . i'll try anything once..... i'm at 94% now on a 16gb class 10...
doesnt seem to be taking as long as some others have......30 mins to an hour ..
edit:.... spoke to soon.. back down to 1% fastfill..hahah
Sometimes it can get kind of hung up. If it does, just close it and start it again. When prompted kill the prior session and it will pick up where it left off. Don't know why it hangs like that but didn't seem to cause problems.
If you have a new card, it is likely not very fragmented yet and will defrag faster.
Mitchell's EVO
Just did this on my Wintech 16gb that is about 85% full. Worked like a charm. My gallery loads almost instantly now, before it took about 20-30 sec. The whole process took about an hour and a half(I plugged my card into my reader in my laptop). Thanks for the heads up on this.
netwokz said:
Just did this on my Wintech 16gb that is about 85% full. Worked like a charm. My gallery loads almost instantly now, before it took about 20-30 sec. The whole process took about an hour and a half(I plugged my card into my reader in my laptop). Thanks for the heads up on this.
Click to expand...
Click to collapse
NP, glad it worked. The whole "SDCards don't need to be defragged" thing is a myth.
Sweet! I'm running it now.
thanks!
thanks for the tip.
paused on me twice. it actually says "finished" (top left corner)
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Can I just use perfectdisk?
quocamole said:
thanks for the tip.
paused on me twice. it actually says "finished" (top left corner)
Click to expand...
Click to collapse
Looking VERY defragged there
freeza said:
Can I just use perfectdisk?
Click to expand...
Click to collapse
If it has an option for defragging flash memory then I suppose so. MyDefrag is free.
Gave this a shot yesterday had to use a card reader and try a couple of times for the software to complete all of the way. My sd card is now Defragmented...did I notice a speed increase...not really but I don't pay much attention to loading times. Makes me feel better and faster and thats good enough for me!
Sent from my PC36100 using XDA App
crap... no mac version
Tried it, took less than 1/2 and noticed a bit of improvement. Sweet.
Not bad...but I just want to say you can defrag without using that program. Connect your phone as a disk drive, then right click it in my computer, properties, the tools, and you can click defrag to analyze and defrag it, and all your other drives.
teh roxxorz said:
Not bad...but I just want to say you can defrag without using that program. Connect your phone as a disk drive, then right click it in my computer, properties, the tools, and you can click defrag to analyze and defrag it, and all your other drives.
Click to expand...
Click to collapse
I was thinking the same thing. In the OP it is explained that this program is designed for flash memory to reduce the number of read/write. Flash memory can only handle so many write cycles so I decided to let this program do it's thing to minimize the amount of moving/writing to my SD card. I am sure that one defrag with any program wouldn't hurt anything but I fell better know that the program tries to minimize the amount it affects a memory cards life cycle.
Just as a side note....I don't know if in the long run it matters a whole lot at all....I can't remember how many times flash memory can be written to but I am sure that I will have a new phone and SD card way before I am in any danger or reaching the upper limits.
teh roxxorz said:
Not bad...but I just want to say you can defrag without using that program. Connect your phone as a disk drive, then right click it in my computer, properties, the tools, and you can click defrag to analyze and defrag it, and all your other drives.
Click to expand...
Click to collapse
Actually, this is incorrect. This program is designed to specifically defrag flash drives. You do not want to defrag them in the same way you defrag typical hard drives.
In the future, please read the OP before commenting. This fact is clearly stated.
Thanks for this! Another great idea from mitchellvii!
backing up Sdcard now will test this
just started defragging my 32gb class 4 lets see what it does.
Q: I kno the OP says around 30 min but was size sd was that? I have around 18gb of music and pictures and videos on my card (love me my music anndd my gf lol) just curious if ill get my phone back before todays kickoff

One XL Rogers LTE Ram size

Hi everyone,
According to XL Rogers specs, I might have 1 gig RAM but I only have 671 Mg.
Is it some king of Rogers limitation?
Can someone having a Rogers One XL release can confirm that?
Annoying because even with a few application in process, Sense is randomly restarting because, I think, of a lack of memory. It often happen right after using a browser for few minutes
Thanks!
I noticed that when looking at ATT One X
Would like to know as well. WTF..
i could be wrong but i think it's because the memory is shared with the gpu.
Weird because it shows 1Gig in parameters/hardware Info.
But still 671 in memory management and benchmark utilities...
That is normal. Some parts of the memory are reserved (for OS, GPU etc.) and are not accessible for all apps. For example, my Galaxy Nexus has 1GB RAM, but only "693 MB" is displayed under "about phone". That has always been the case for all Android phones, although how much of the memory is reserved differs between phones/ROMs.
tsekh501 said:
That is normal. Some parts of the memory are reserved (for OS, GPU etc.) and are not accessible for all apps. For example, my Galaxy Nexus has 1GB RAM, but only "693 MB" is displayed under "about phone". That has always been the case for all Android phones, although how much of the memory is reserved differs between phones/ROMs.
Click to expand...
Click to collapse
I wonder why this always comes up when a new phone is released. When people buy a computer with 6 gigs of ram, do they get mad when Windows or OSX uses 2 of it?
It's the same concept here, but this always comes up.
If you run:
cat /proc/meminfo
That should give you all the physical memory available to Android.
Squeak22 said:
I wonder why this always comes up when a new phone is released. When people buy a computer with 6 gigs of ram, do they get mad when Windows or OSX uses 2 of it?
It's the same concept here, but this always comes up.
Click to expand...
Click to collapse
It's not really the same concept, as that has to do with the 32-bit limitation. And yes, people did complaining a lot a few years ago when 4GB RAM started to become really affordable. People rarely complained about that these days, not because they finally understood the issue, but rather most computers are shipped with 64-bit OS now.
tsekh501 said:
It's not really the same concept, as that has to do with the 32-bit limitation. And yes, people did complaining a lot a few years ago when 4GB RAM started to become really affordable. People rarely complained about that these days, not because they finally understood the issue, but rather most computers are shipped with 64-bit OS now.
Click to expand...
Click to collapse
/facepalm
When you load up task manager in windows, it shows how much free memory you have. The amount of free memory is NEVER what your total physical memory is. There is always about a third used for cached data, OS processes, and stupid widgets that you have running on the desktop.
Nowhere did I mention the 32bit limitation, that would be a valid issue.
Squeak22 said:
/facepalm
When you load up task manager in windows, it shows how much free memory you have. The amount of free memory is NEVER what your total physical memory is. There is always about a third used for cached data, OS processes, and stupid widgets that you have running on the desktop.
Nowhere did I mention the 32bit limitation, that would be a valid issue.
Click to expand...
Click to collapse
There's no need to be so rash. He does have a valid point.
cat /proc/meminfo returns the following:
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Whereas the task manager displays 672MB total memory (used+free):
danada said:
There's no need to be so rash. He does have a valid point.
Click to expand...
Click to collapse
150ish meg for ramdisk and another 150ish meg for GPU. Explained away.
This is common on every phone that's ever come out for the last year + with 1 gigs of ram.
Squeak22 said:
/facepalm
When you load up task manager in windows, it shows how much free memory you have. The amount of free memory is NEVER what your total physical memory is. There is always about a third used for cached data, OS processes, and stupid widgets that you have running on the desktop.
Nowhere did I mention the 32bit limitation, that would be a valid issue.
Click to expand...
Click to collapse
That's free memory that you are talking about. Its total physical memory and I and OP were saying. My Windows 7 task manager always shows the actual total physical memory, and that wasn't the case with winxp before either sp2 or sp3 and tons of people whined about that on a daily basis.
The OP's question, however, concerns Android and it doesnt show the correct amount of physical memory on the phone. He is not saying "free memory", which is of course changing all the time depending on running apps.
Sent from my Galaxy Nexus using XDA
I am seeing a total of 970ish MB's on my international htc one x. Currently 709MB used and 267MB free.
I believe some of the RAM is reserved for the GPU and since there are different GPU's in the two models it is probably what makes up the difference in available RAM.

[INFO] All about Memory Cards

Ok people, firstly, forget the thread title. Wrote it that way just to catch your eye
Anyways coming to the point. I've seen in many threads (especially Sense Rom threads) about questions and discussions about Size and Class of SD card.
They were all about buying cards with higher class for better speed and for providing a lag free experience.
Even I was considering buying a new one, so did some research [by research, I mean lot of Googling ].
Decided to share what I learned and in the process help you in your choice.
Trying to be as simple as possible, even to prove helpful to newbies.
Click to expand...
Click to collapse
WHAT ARE SD CARDS?​
Well SD stands for Secure Digital. It is one of the formats of Memory Card.
This format includes 4 types of cards.
The Standard Capacity (SDSC)
The High Capacity (SDHC)
The Extended Capacity (SDXC)
The SDIO (which combines input/output functions with data storage)
And these types come in basically 3 different sizes - Original, Mini and Micro.​​
Since now we know what are SD cards, let's move on to the Speed Class Rating of a SD card.
What does the Class of an SD card reveal?​We usually talk about the speed of cards in terms of 'Read' and 'Write' speeds.
But while assigning a Class to a SD card, only the Write speeds of that particular SD card are considered.
With SD cards, manufacturers generally rate the cards in three different ways.
The first method is through the use of Speed Class Ratings.
For example, most SD cards will be rated at Class 2, Class 4, Class 6 or Class 10.
The number denotes the minimum sustained Write speed for an empty card. So real-time speeds generally vary and are usually higher.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
The second method is a newer method of the class ratings called UHS (Ultra High Speed). UHS-I cards are much, much faster and can support up to 104 MB/s.
The third method is through the use of “x” ratings. For most computer enthusiasts, this should be quite familiar since it’s just a multiple of 150kB/s, the speed of the standard CD-ROM drive (Oh yeah, that old thing…). Therefore, an SD card with a 133x speed rating should be able to achieve a transfer rate of 19,950 kBytes/s, which is about 20 MB/s. High speed SD cards will sometimes use the x rating as class ratings end at Class 10.
​However there's a catch.
The Write speeds used to denote a Class are "Sequential Write Speeds". Well you might ask, whether it matters to a normal mobile user.
The answer is maybe or may not.
Here's why: The Sequential Write speeds or Sequential Access is the speed with which large amount of data of a SINGLE FILE is written.
For example, while transferring data from system to your phone or vice versa, or when a digital camera records large photographs or videos.
However, for a smartphone user, there is another important speed to be considered - "Random Access". This is writing a large number of small files. Random read-write is the most important for speed during normal phone performance.
One study found that, in this Random-Access use, some Class 2 cards achieved a write speed of 1.38 Mbit/s, while all cards tested of Class 6 or greater (and some of lower Classes) including those from major manufacturers, were over 100 times slower.
By saying that, I'm not suggesting that all Class 2 SD cards are good for your phone.
You might wonder, why does Higher Class cards get lower Random Access readings?
Class 2 and 4 cards have better Random Access speed and Random read/write than class 6 or 10 cards in general. To get higher Sequential Write speeds, manufacturers sacrifice the Random Access speeds.
That's just how it is.
Class 6 and 10 cards are setup to initialize the area of the card that is about to be used to make the sequential write/read quicker, but it takes extra time in the beginning to do the initialization. If your using the card where you need to access small bits of info from various parts of card quickly, you will do best with class 2 or 4.
Most people seem to assume that the higher the class, the better the card is for everything, but that's not how it is. Different purposes need different kinds of speed... cards with higher sequential speeds have lower random speeds
Click to expand...
Click to collapse
.
GEEK ALERT: Before you read further, you must be a geek or newbie who is greedy to grab everything that's thrown at him.
If you're just looking for the nutshell, ignore what's given below.
Given below are bunch of benchmarks that usually are to be taken into consideration, rather than just "Read Speed and Write Speed"
Sequential Read
Sequential Write
Random Read 512KB
Random Write 512KB
Random Read 4KB (QD=1)
Random Write 4KB (QD=1)
Random Read 4KB (QD=32)
Random Write 4KB (QD=32)
How to benchmark your card?
​Benchmarks are a rather elusive science since there are too many variables that can impact the results, not the least of which is the card itself. Anyone who has worked with SD cards in the real world has probably noticed that the brand of card can make a dramatic difference in performance. Tests have shown as much as a 12x difference in performance between best and worst cards in real world tasks. As already noted, the "class" ratings used by manufacturers as a measure of performance are not reliable indicators of performance either. In fact, some manufacturer's class 2 and class 4 rated cards routinely outperform cards rated as class 6 or class 10!
Consistency of testing is important since different devices and bottlenecks, such as bus speeds, buffers, file systems and activity can all impact and alter the results. This is why a card that has performed extremely well on a test may turn out to be a dog in your device. Short of testing all cards in a lab environment under exacting circumstances may well yield inconsistent results.
​There are a lot of apps available for bench-marking, for Android, as well as Windows and Linux.
Bench-marking Tools​
CrystalDiskMark
HD Tune
AndroBench
Others such as SD Tools, Antutu etc.
​
Few Examples that might help:
Now we come to the question - Which SD cards should we go for?
Well, that's upto you, the end user, to decide based on your requirements.
On my part, I've provided you the information to help you with your decision.
Yet, for those who don't like to take decisions, here's something which might help.
The SD Association defines standard speed classes indicating minimum performance to record video. Both read and write speeds must exceed the specified value. These are defined in terms of suitability for different applications:[40]
Class 2 for SD video recording
Class 4 and 6 for high-definition video (HD) to Full HD video recording,
Class 10 for Full HD video recording and consecutive recording of HD stills
UHS Speed Class 1 for real-time broadcasts and large HD video files.
Click to expand...
Click to collapse
Still confused?
Well I'll let those who already bought new SD cards to post their reviews.
Source: Wikipedia, CustomPCReview, Kunal'sBlog, XDA and many others which I don't remember.
Click to expand...
Click to collapse
Just in case
You never know!
Nice sharing ..............!! Very informative
:good:
Add how to un partition sdcard also (Ubuntu)
Sent from my HTC Explorer using xda premium
Sd card sandisk class 10 (sdhc)
[I have bought a new class 10 SD card of Scan Disk, but as per told but DEV's that using class 10 SD card on Sense 4, there are no lags and no random reboots, but still i face lags and random reboots, i dont i have seen any difference in class 4 or class 10 SD card.
Can u let me know if i am doing something wrong in making the partition on the card.
Nice theoretical information doc.
Sent from my HTC Explorer A310e using xda premium

[APP][2.2+] A1 SD Bench

A1 SD Bench · Free
Free tool to benchmark your phone & tablet SD card, RAM and memory, to test IO read / write speed (Quick, Long, Random I/O and Accurate) and check media performance, and compare your benchmark results online.
Download A1 SD Bench from Google Play.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
New v2.4.0 (Dec 15, 2014):
New: Option to hide ADs
Improved: More devices are recognized
Improved: Misc improvements
Features:
Free
Four benchmark modes - Quick, Longer, Accurate and Random I/O
Quick and Accurate benchmark (to reduce caching effects)
Built-in compare results online (anonymous)
Built-in view statistics online
SD card speed test, or multiple SD cards
Speed test any Android storage including RAM and USB
Measure both read and write speeds
Speed categories: Class 2 .. Class 10; UHS-I, UHS-II and UHS-III
Custom location option to test any device directory / location
Shows total and free memory for all media types including RAM
Supported devices:
Internal memory
Internal / External SD cards
System RAM
USB flash drive
USB HDD (Hard drives)
Up to 6 USB devices (connected simultaneously via USB hub)
Notes:
Problem? Please, send us your feedback using Send Feedback feature from the Settings screen
We can not test all Android devices, so if your device's memory / SD card / USB is not detected correctly, use Send Feedback feature from the Settings screen, specifying path to your storage, e.g. /mnt/sdcard/, /mnt/usb_drive;
NTFS and other non-native file systems is supported as far as your Android device is compatible with a such FS.
Contact:
From the application itself
web feedback form
Post in this thread
Nice!!!
Nice but antutu is best Benchmark andriod app forever!!!!!!
I used this app so much when i had my galaxy s2. It's a great app and a must have to any toolbox.
Sent from my Nexus 5 using Tapatalk
Added link
A1:dev said:
Post in this thread
Click to expand...
Click to collapse
Try adding link - https://play.google.com/store/apps/details?id=com.a1dev.sdbench
anu.cool said:
Try adding link...
Click to expand...
Click to collapse
Thanks for the link. Unfortunately, I can not add a links yet, due to new membership.
A1:dev said:
Thanks for the link. Unfortunately, I can not add a links yet, due to new membership.
Click to expand...
Click to collapse
click on mutiquote on my msg, then go to editting of ur first post,go advanced edit, then u will see a new msg below where u type the msg, it will prompt u to add the quoted msg, click yes, then save ur post.... i think tht shud help u add link to ur orignal post ..
anu.cool said:
click on mutiquote on my msg, then go to editting of ur first post,go advanced edit, then u will see a new msg below where u type the msg, it will prompt u to add the quoted msg, click yes, then save ur post.... i think tht shud help u add link to ur orignal post ..
Click to expand...
Click to collapse
Thanks... however "it will prompt u to add the quoted msg" did not happened
Still receiving "To prevent spam on the XDA forums, ALL new users prevented from posting outside links in their messages. After approximately 10 posts, you will be able to post outside links. Thank you for understanding!"
Dear developer, I'm using an Anycall brand Class 10 UHS-I micro sd card on a non-rooted Note 3. When trying to run the benchmark test, it reported error that target file can not be found (permission denied.) Any clue how I can fix it? Thanks!
wildbizzar said:
Dear developer, I'm using an Anycall brand Class 10 UHS-I micro sd card on a non-rooted Note 3. When trying to run the benchmark test, it reported error that target file can not be found (permission denied.) Any clue how I can fix it? Thanks!
Click to expand...
Click to collapse
+1
wildbizzar said:
Dear developer, I'm using an Anycall brand Class 10 UHS-I micro sd card on a non-rooted Note 3. When trying to run the benchmark test, it reported error that target file can not be found (permission denied.) Any clue how I can fix it? Thanks!
Click to expand...
Click to collapse
In a next few hours new version will be available in Google Play. If you have KitKat 4.4 device, please test and provide your feedback. At this point, we can not test 4.4 on real device.
Change log v2.1.0:
• New: Android 4.4 KitKat support (Beta version; please send your feedback)
• Improved: More devices are recognized
• Improved: Added "Strontium Nitro" SD card
• Changed: Default installation location is now set to "auto" instead of "SD card" (sorry Android 2.3 folks)
A1:dev said:
In a next few hours new version will be available in Google Play. If you have KitKat 4.4 device, please test and provide your feedback. At this point, we can not test 4.4 on real device.
Change log v2.1.0:
• New: Android 4.4 KitKat support (Beta version; please send your feedback)
• Improved: More devices are recognized
• Improved: Added "Strontium Nitro" SD card
• Changed: Default installation location is now set to "auto" instead of "SD card" (sorry Android 2.3 folks)
Click to expand...
Click to collapse
Thanks a lot. Now it works like a champ. Appreciate your quick support on the issue!
New version (v2.2.0)
New v2.2.0 (Aug 11, 2014):
New: Initial support for “Android L (Preview)”; fixes FC
Fixed: Detail view of frequencies now shows Hz correctly
Fixed: Custom location enabling/disabling caused false UI glitch
Improved: Supports many more USB and OTG devices
Improved: Added G.Skill SD cards
Improved: More devices are recognized
Improved: Misc improvements
Changed: Internal and External cards are shown simply as SD Card
New version (v2.4.0)
We have just released a new version. Available at Play Store.
Got a question on how the benchmark works:
I've got three different brands of SD cards here: Kingston SDCA3 32GB, Transcend Ultimate 633x 32GB and Samsung Pro 64GB. All cards have roughly the same speed rating of W80 and R90. And when it comes to read speeds they all get well over 80MB/s, so that's fine.
When it comes to write speeds however the two 32GB models only achieve about 40+MB/s, whereas the 64GB model gets well over 70MB/s.
Why is that?
All cards are exFAT formatted with 128 kB cluster size and changing it to 32 kB doesn't seem to have any effect.
Phone is an LG G3 16GB D855 running Android 5.0.
Any help or information is appreciated.
Thanks in advance
An-Druid said:
I've got three different brands of SD cards here: Kingston SDCA3 32GB, Transcend Ultimate 633x 32GB and Samsung Pro 64GB.
When it comes to write speeds however the two 32GB models only achieve about 40+MB/s, whereas the 64GB model gets well over 70MB/s.
Why is that?
Click to expand...
Click to collapse
May be a little late answer, but anyway... Card class or rating does not mean anything. It usually only guarantees the minimum speed. The maximum speed greatly varies between cards and manufacturers. Also, not all devices have bus/card reader that can achieve maximum performance of the card.
An-Druid said:
All cards are exFAT formatted with 128 kB cluster size and changing it to 32 kB doesn't seem to have any effect.
Click to expand...
Click to collapse
Cluster should affect Random read/write speed... And sequential usually is almost unaffected. But again, different devices, different cards can produce different results.

[APP] [ROOT] [Android 4.0+] [v 1.0.7] SSD Boost

Hi, I'm Andrea, a new member in XDA forum,
I'd like to introduce to you my app 'SSD Boost'
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
DOWNLOAD​
This is a work in progress project which has the goal to speed performance and extend endurance of smartphone internal flash memory.
Smartphone internal storage uses solid state technology SSD/emmc. Counterpart to lower access time, and less latency of SSD is that its performance may degrade over time: writing after writing, the memory cells show a measurable drop in performance, and will continue degrading throughout the life.
'SSD Boost' minimizes writings on your SSD, increasing speed and extending life of your storage.
Though the main goal of the app is the optimization of flash memory, many users have also experienced a better responsiveness in the startup of the apps, a longer battery endurance and a decreased temperature of the phone.
Probably because 'SSD Boost' app is a workaround to some insane process running in the background, that keeps writing to disc. This has been experienced from Android ICS version to Nougat (both 7.1.1 stock and CM 14.1).
It needs root.
Be careful, if you have got one of the following models:
- Samsung Galaxy R.
- Samsung Galaxy SII.
- Samsung Galaxy Note.
- Samsung Galaxy Nexus.
- Samsung Epic 4G Touch.
before installing, check presence of Brickbug (a bug on eMMC chip fw): download 'eMMC Brickbug Check' from Play Store
Changelog:
1.0.7
- improved root check method
1.0.2
- added TRIM command
What does TRIM for ssd/nand performance.
Unlike HDD, SSD cannot overwrite data because only a transition from 0 to 1 is possible, so before overwriting the data has to erase the whole page to 0 then program the bits.
Erasing takes longer then writing data, so after you used the device for a while it will have to erase the pages before writing them and cripple performance.
It would be so much better if controller would erase the blocks while idle so you have them ready when you need to write. Unfortunately this is not possible as controller has no knowledge of file system so it does not know what blocks are used and with are not.
TRIM command, when supported by the SSD/NAND, allows operating system to send a list of blocks to controller to erase when is idle.
Android 4.3+ supports trim natively, it means that operating system after deleting a file will sends a trim command to nand controller notifying witch blocks the file used, the controller will erase them when idle so it can write fast without erasing next time.
But users have no control over this process and Android doesn't trim the storage for a very long time.
'SSD Boost' helps you to run trim on your device manually if you feel it has become laggy and anyway it does trim for you on every reboot.
1.0.1
- added NOATIME, NOADIRATIME support
Android maintains file system metadata that records when each file was last accessed. This timestamp is known as atime and atime comes with a performance penalty – every read operation on a filesystem generates a write operation.
Updating the atime every time a file is read causes a lot of usually-unnecessary IO, slowing everything down.
'SSD Boost' disables the tracking of atime, remounting filesystems with the noatime/noadiratime options.
You can freely download the latest version from Play Store
Enjoy it, and give me a feedback.
Andrea
/reserved
/reserved
Quick question: what exactly would the risk of this ****ing up my phone be?
Quick answer: If you have the patience to read, I wrote about some line above.
thanks for the work
do we need to run it regularly?
lm.silva said:
thanks for the work
do we need to run it regularly?
Click to expand...
Click to collapse
It's not necessary, it autostarts on boot.
You can find the new release 1.0.8 on Play Store.
This is this app https://play.google.com/store/apps/details?id=com.puandr.boost ?
Devhost97 said:
This is this app https://play.google.com/store/apps/details?id=com.puandr.boost ?
Click to expand...
Click to collapse
Yesss, thank you
I have released on Play Store an alpha version with increased read cache features.
Who wants in advance to test this version, contact me in pvt, I'll be giving him/her free Pro features upgrade.
Thanks
I like what I read about your app! Cool idea. How exactly do you limit the useless reads/writes? How do you know which ones to block or limit?
If you will make a open beta tests i can join
Is this Thang still on?
mahirozdin said:
If you will make a open beta tests i can join
Click to expand...
Click to collapse
Yes, there is already an open beta test group
Really, there is also an alpha test group, just contact me in pvt to join.
revized said:
Is this Thang still on?
Click to expand...
Click to collapse
Yes.
Cool going to get the pro version to keep this lag off my home base
On 4th August 2007, Linux kernel developer Ingo Molnár called atime "perhaps the most stupid Unix design idea of all times," adding: "[T]hink about this a bit: 'For every file that is read from the disk, lets do a ... write to the disk! And, for every file that is already cached and which we read from the cache ... do a write to the disk!'"
"Atime updates are by far the biggest IO performance deficiency that Linux has today. Getting rid of atime updates would give us more everyday Linux performance than all the pagecache speedups of the past 10 years"
The problem is that, ten years later, Android, based on Linux kernel, still uses atime updates.
How to solve it, then?
The first idea is, as a workaround, to remount Android filesystems with the noatime/noadiratime options.
But which impact can this produce to apps that use this timestamp file system metadata?
I actually don't exactly know. For this reason I've developed an experimental app (this one ) that does exactly this, it remounts filesystems.
Nowadays, this app has been installed on more than 100.000 Android devices, so it's possible to get some feedbacks.
(to be continued)
Wow man , just installed it on my Galaxy Note 2 with crDroid 7.1.2 ROM. Phone is working faster and there is no lag at all , like it was before , especially while using keyboard or scrolling pages!
Thank you nNeFeroN.
I just received the following review:
"After i run this app it works so fast ram usage goes low but after i go for data managing in a charger usb cable that i plugged up in my phone it doesn't read anything even notifications of the (Connected As) Media/Audio Doesn't shows up or it didn't work cuz i trimmed my SDD?.but i'm a little bit happy cuz ur app helped me too."
He is using a Samsung Galaxy Grand Prime (fortunave3g), Android 4.4

Categories

Resources