Jump to content











Photo

Is Transcend Premium 300x UHS-I card overrated?


  • Please log in to reply
55 replies to this topic

#1 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 04:37 AM

I plugged 32GB Transcend microSDHC Class 10 UHS-I 300x (Premium) card (TS32GUSDU1) in Transcend USB 3.0 Card Reader (TS-RDF5). Since both the card and the reader are UHS-I compatible, I decided to test the rated capability of the card. Also, the card is supposed to deliver its full potential when used in UHS-I host.


Product: 32GB Transcend microSDHC Class 10 UHS-I 300x (Premium) card (TS32GUSDU1)

 

Size 11mm x 15mm x 1mm Op. Voltage 2.7V~3.6V Op. Temperature - 25°C(- 13°F)~85°C(185°F) Durability 10,000 insertion/removal cycles Weight 0.4g Transfer Rate (Max)* 45MB/s 300x *Note Speed may vary due to host hardware, software and usage.



Product: Transcend USB 3.0 Card Reader (TS-RDF5)


Dimensions 56mm (L) × 24mm (W) × 8.9mm (H) Weight 18g Interface USB 2.0 / USB 3.0 Slot Power Output 5V Operating Temperature 0°C (32°F)~70°C (158°F) Certificates CE, FCC, BSMI, C-tick, KC Supported Cards SDHC (UHS-I), SDXC (UHS-I), microSDHC (UHS-I), microSDXC (UHS-I) System Requirements Desktop or notebook computer with USB port (USB3.0 port needed for USB3.0 performance) One of the following Operating Systems:

  • Windows XP
  • Windows Vista
  • Windows 7
  • Windows 8
  • Mac OS 10.2.8 or later
  • Linux Kernel 2.6.30 or later

 


My laptop is running Windows 7 x64 Ultimate SP1 on Intel Core i3 2.27 GHz equipped with 4GB DDR3 RAM and USB 2.0 ports. Benchmarking software was CrystalDiskMark 3.0.3 x64. USB 2.0 can ONLY limit the speed beyond 35 MB/s. According to Wikipedia,

 

USB 2.0 was released in April 2000 (now called "Hi-Speed"), adding higher maximum signaling rate of 480 Mbit/s or 60 MB/s (due to bus access constraints the effective throughput is limited to 35 MB/s or 280 Mbit/s), in addition to the "USB 1.x Full Speed" signaling rate of 12 Mbit/s

USB 3.0 was released in November 2008. The standard defines a new SuperSpeed mode with a signaling speed of 5 Gbit/s (625 MB/s) and, due to encoding overhead, usable data rate of up to 4 Gbit/s (500 MB/s).


Now, the worse part comes. Since TS-RDF5K supports UHS-I, I presume it acts as an UHS-I host by their definition. Though the card is rated to have a maximum speed of 45 MB/s if UHS-I host is used, in reality it offered 14.6 MB/s, which is way less than what is advertised. Can't blame the usage of USB 2.0 as the practical maximum is 35 MB/s.

Test system: Intel Core i3 2.27 GHz, Windows 7 Ultimate x64, USB 2.0 host, Transcend TS-RDF5 USB 3.0 Adapter, 4 GB DDR3

TS_RDF5.jpg


Test system: Intel Core i3 1.80 GHz, Windows 8.1 x64, USB 3.0 host, Transcend TS-RDF5 USB 3.0 Adapter, 4 GB DDR3

Transcend_Premium_on_3_0.jpg



#2 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 05:46 AM

No, I don't think it's overrated. It has a quite good 4kB-random-write and the other numbers are not surprising, considering it's a micro-SDHC, which is a very limiting form factor. Of course, I'd expect more of a full-sized SDHC card, and even more of a pendrive. But if you compare it to the results of my pendrive tests (which I attached to one of my posts in the pendrive thread you started), you can see it's a good enough (for 2014) micro-SDHC card you have there.



#3 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 06:05 AM

No, I don't think it's overrated.

 

Isn't it? It has been rated to be a 45 MB/s card when used in UHS-I host. It has under-beaten my expectation by offering less than 1/3 speed of its full potential. The tests I threw at it used ~1GB data. It should be pretty enough to realize its maximum potential.

 

 

It has a quite good 4kB-random-write and the other numbers are not surprising, considering it's a micro-SDHC, which is a very limiting form factor. Of course, I'd expect more of a full-sized SDHC card, and even more of a pendrive. But if you compare it to the resluts of my pendrive tests (which I attached to one of my posts in the pendrive thread you started), you can see it's a good enough (for 2014) micro-SDHC card you have there.

 

I agree, sequential write is what I was hurt at.



#4 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 07:59 AM

Even 35 MB/sec is not reached on USB 2.0, except on very select machines, with the fastest possible processors and southbridges and a great USB 2.0 controller on the adapter. Moreover, as in quite a lot of other natural processes, the maxing-out is exponential, not linear. Bearing that in mind, I'd say your test shows maxed sequential-read and 512 k-read, and you'd need a very good card to go above what you're getting. Now, the sequential-write is really not great. And, of course, writes (of any kind) are always more exacting processes. But then, you should ask yourself why a 32 GB SanDisk Extreme Class 10 UHS-I, also rated at 45 MB/sec, costs about twice as the Transcend... Below is some data for you to muse over. The card is a full-size SDHC, not a micro-SDHC, but I used the same card in both tests... all I wished to highlight is that different adapters max-out at different maximum tranfer rates, and that they do affect the results even well below the maximum, too.
 
--------------------------------------------------
SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
Adapter: Kingston MobileLite G3 (FCR-MLG3) on USB 2.0
--------------------------------------------------
CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
      Crystal Dew World : http://crystalmark.info/
--------------------------------------------------

   Sequential Read :   30.495 MB/s
  Sequential Write :   26.600 MB/s
 Random Read 512KB :   29.218 MB/s
Random Write 512KB :   15.107 MB/s
   Random Read 4KB :    4.505 MB/s
  Random Write 4KB :    2.107 MB/s

         Test Size : 100 MB
--------------------------------------------------
SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
Adapter: Transcend RDS-5 on USB 2.0
--------------------------------------------------
CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
      Crystal Dew World : http://crystalmark.info/
--------------------------------------------------

   Sequential Read :   19.523 MB/s
  Sequential Write :   17.292 MB/s
 Random Read 512KB :   19.258 MB/s
Random Write 512KB :   11.842 MB/s
   Random Read 4KB :    4.611 MB/s
  Random Write 4KB :    2.321 MB/s

         Test Size : 100 MB


#5 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 08:20 AM

Moreover, as in quite a lot of other natural processes, the maxing-out is exponential, not linear.

 

Means?

 

 

 

Bearing that in mind, I'd say your test shows maxed sequential-read and 512 k-read, and you'd need a very good card to go above what you're getting.

 

Actually, it's supposed to be a very good card. Have a look at its reviews

 

 

 

But then, you should ask yourself why a 32 GB SanDisk Extreme Class 10 UHS-I, also rated at 45 MB/sec, costs about twice as the Transcend

 

It's not. Sandisk Extreme Class 10 UHS-I costs 1689 INR while Transcend Class 10 UHS-I costs 1250 INR. This much difference in price does not anyway mean that Sandisk card is better.



#6 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 08:47 AM

Yes, it does. The SanDisk *is* better. But the differential in price is bigger in Brazil (as always).
Give a good look at the tests I added to my previous post. The card is maxing-out both adapters, but the MobiliLite G3 maxes (on USB 2.0, of course) at about 30 MB/sec (it's the southbridge USB controller that's actually maxing out, in this case), while the RDS-5 maxes out at about 20 MB/sec, regardless of the card. That is part of what I meant by remarking the maxing-out is exponential, not linear... I'll search for a graph expressing that and post it for you later (unless Wonko finds it way before I do, of course...)

BTW, at amazon.com (in the USA):

SanDisk Extreme 32GB MicroSDHC UHS-1 Speed Up To 45MB/s With Adapter - US$27.45

Transcend 32GB MicroSDHC Class10 UHS-1 Speed Up To 45MB/s With Adapter - US$17.99

#7 Wonko the Sane

Wonko the Sane

    The Finder

  • Advanced user
  • 16066 posts
  • Location:The Outside of the Asylum (gate is closed)
  •  
    Italy

Posted 14 June 2014 - 10:11 AM

Wonko can find one alright (of course ;)) but are you sure you are going "exponential" (and not - just saying - "polynomial", "logarithmic" or "power") ?

http://excel.officet...ning/trendlines

 

:duff:

Wonko



#8 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 10:51 AM

The following two tests differ only in the "Test Size" and highlights how the sequential write values tend to approach the sequential read values as the test-file size gets bigger, as expected for something that approaches more and more an asymptote:

    --------------------------------------------------
    SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
    Adapter: Dell U2410F Monitor Card-Reader on USB 2.0
    --------------------------------------------------
    CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
          Crystal Dew World : http://crystalmark.info/
    --------------------------------------------------

       Sequential Read :   20.888 MB/s
      Sequential Write :   16.847 MB/s
     Random Read 512KB :   20.503 MB/s
    Random Write 512KB :    9.598 MB/s
       Random Read 4KB :    4.393 MB/s
      Random Write 4KB :    2.073 MB/s

             Test Size : 100 MB <<<

    --------------------------------------------------
    SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
    Adapter: Dell U2410F Monitor Card-Reader on USB 2.0
    --------------------------------------------------
    CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
          Crystal Dew World : http://crystalmark.info/
    --------------------------------------------------

       Sequential Read :   20.591 MB/s
      Sequential Write :   19.397 MB/s
     Random Read 512KB :   20.527 MB/s
    Random Write 512KB :    9.647 MB/s
       Random Read 4KB :    4.397 MB/s
      Random Write 4KB :    2.146 MB/s

             Test Size : 1000 MB <<<

I think perhaps "logistic" is probably more what I had in mind... but anyway my point was "linear" (green) shoots straight into the limit (blue), while exponential (red) goes asymptotically towards it. Something like this:

 

post-non-linear.gif



#9 Wonko the Sane

Wonko the Sane

    The Finder

  • Advanced user
  • 16066 posts
  • Location:The Outside of the Asylum (gate is closed)
  •  
    Italy

Posted 14 June 2014 - 11:12 AM

"Logistic"? :unsure: That is seemingly a "logarithmic" trend, but if you invert X and Y, it may also be exponential (either strictly exponential or "natural exponential" in base e) and it could also be (of course) a suitable polynomial.

See:

http://www.tushar-me...analysis/16.htm

for an "all in one" image:

image003.gif

 

:duff:

Wonko



#10 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 11:34 AM

Give a good look at the tests I added to my previous post. The card is maxing-out both adapters, but the MobiliLite G3 maxes (on USB 2.0, of course) at about 30 MB/sec (it's southbridge maxing out actually), while the RDS-5 maxes out at about 20 MB/sec, regardless of the card.


Kingston MobileLite G3 (FCR-MLG3) is an USB 3.0 card reader while Transcend RDS-5 is an USB 2.0 one. One which I tested with is Transcend RDF-5 which is again an USB 3.0 product.

 

Also we can't really say whether Kingston one is being maxed out as we are lacking in an evidence establishing the fact that the card is, indeed, no slower (if not faster) than the reader itself.



#11 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 11:46 AM

Right! So: the RDF-5, just like the MobileLite G3, should not contribute to the maxing, because they have USB 3.0 speed capable controllers, and hence one sees 30-32 MB/sec (with a 100 MB test file size). Now, the (Alcor AU6336) controller in the RDS-5 limits it to 19-21 MB/sec max, instead. The non-linearity shows itself more evidently when comparing the 512k-random-write values (if things were linear they should be about the same...).

#12 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 11:55 AM

Let me summarize, in due course of discussion, we are trying to find answers to the following questions:

  1. Is the capacity (45 MB/s) of 32GB Transcend microSDHC Class 10 UHS-I 300x (Premium) card (TS32GUSDU1) overrated?
  2. Does Transcend USB 3.0 Card Reader (TS-RDF5).make use of ANY potential of USB 3.0?
  3. Does Transcend USB 3.0 Card Reader (TS-RDF5) truly utilize the full potential of an UHS-I card or does "being UHS-I compliant" merely mean "being able to read/write" those cards?
  4. How can we explain the 1/3 of rated speed exhibited in the benchmark results in my original post? In other words, which one is capping the speed - card/card reader or both?


#13 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 12:02 PM

Right! So: the RDF-5, just like the MobileLite G3, should not contribute to the maxing, because they have USB 3.0 speed capable controllers, ...

 

C'm on. Neither your nor mine uses USB 3.0 host port, hence readers defaults to USB 2.0 as per specification.



#14 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 12:04 PM

Here's USB 3.0... Please do notice the three smaller values are just about the same (pairwise) on either USB. When the controller enters "USB 2.1 mode" it merely falls back to using four instead of 9 pins for communication, using USB 2.0 signaling and protocols, and limiting its current demands to no more than 500 mA (preferably much less than that).
 
--------------------------------------------------
SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
Adapter: Kingston MobileLite G3 (FCR-MLG3) on USB 3.0
--------------------------------------------------
CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
      Crystal Dew World : http://crystalmark.info/
--------------------------------------------------

   Sequential Read :   46.186 MB/s
  Sequential Write :   35.286 MB/s
 Random Read 512KB :   42.869 MB/s
Random Write 512KB :   16.641 MB/s
   Random Read 4KB :    5.007 MB/s
  Random Write 4KB :    2.097 MB/s

         Test Size : 100 MB


#15 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 12:18 PM

--------------------------------------------------
SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
Adapter: Kingston MobileLite G3 (FCR-MLG3) on USB 3.0
--------------------------------------------------
CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
      Crystal Dew World : http://crystalmark.info/
--------------------------------------------------

   Sequential Read :   46.186 MB/s
  Sequential Write :   35.286 MB/s
 Random Read 512KB :   42.869 MB/s
Random Write 512KB :   16.641 MB/s
   Random Read 4KB :    5.007 MB/s
  Random Write 4KB :    2.097 MB/s

         Test Size : 100 MB

 

 

--------------------------------------------------
SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
Adapter: Kingston MobileLite G3 (FCR-MLG3) on USB 2.0
--------------------------------------------------
CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
      Crystal Dew World : http://crystalmark.info/
--------------------------------------------------

   Sequential Read :   30.495 MB/s
  Sequential Write :   26.600 MB/s
 Random Read 512KB :   29.218 MB/s
Random Write 512KB :   15.107 MB/s
   Random Read 4KB :    4.505 MB/s
  Random Write 4KB :    2.107 MB/s

         Test Size : 100 MB
--------------------------------------------------
SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB
Adapter: Transcend RDS-5 on USB 2.0
--------------------------------------------------
CrystalDiskMark 2.2 (C) 2007-2008 hiyohiyo
      Crystal Dew World : http://crystalmark.info/
--------------------------------------------------

   Sequential Read :   19.523 MB/s
  Sequential Write :   17.292 MB/s
 Random Read 512KB :   19.258 MB/s
Random Write 512KB :   11.842 MB/s
   Random Read 4KB :    4.611 MB/s
  Random Write 4KB :    2.321 MB/s

         Test Size : 100 MB

 

 
Armed with the data above, what we can conclude are
  • SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB, indeed, maxed out Transcend RDS-5 on USB 2.0
  • SANDISK Extreme (45 MB/s, Class 10, UHS-I) 32 GB is having quite near to its rated capacity, provided used on USB 3.0
Unfortunately, following comparisons are still impossible

To comment on above, swapping of the cards/adapters is needed.



#16 Wonko the Sane

Wonko the Sane

    The Finder

  • Advanced user
  • 16066 posts
  • Location:The Outside of the Asylum (gate is closed)
  •  
    Italy

Posted 14 June 2014 - 12:20 PM

OT :w00t:, but not much ;), what may happen with those (BTW stupidly small) microSDHC may :ph34r: be not completely unlike what has been recently documented for CF cards:
http://www.msfn.org/...act-flash-card/

:duff:
Wonko



#17 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 12:43 PM

But now we can infer with reasonable certainity that the disappointing 14.58 MB/sec seq. write value you measured is just about the true value for that card on either USB (and the same holds for all the measured values smaller than 14.58 Mb/sec, too). So that, as I said before, seems to me to be just about right, for a Transcend. Please bear in mind that the 45 MB/sec SanDisks are EoL... current ones are rated 80 MB/sec for the micro-SDSC and 95 MB/sec for the full-sized cards, nowadays.

#18 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 12:54 PM

But now we can infer with reasonable certainity that the disappointing 14.58 MB/sec seq. write value you measured is just about the true value for that card on either USB. So that, as I said before, seems to me to be just about right, for a Transcend.

 

There are other unfortunate possibilities, too  :dubbio:

  • The card is a fake one and delivering less than rated speed
  • The reader is a fake one and maxed out at an unacceptable speed
  • The card is shamelessly overrated
  • The reader contains a worse than average 3.0 controller and maxed out at an unacceptable speed
  • The reader is unable to exercise full potential of UHS-I cards


#19 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 14 June 2014 - 01:34 PM

Here are two noteworthy reviews from amazon.com's customers, for that Transcend 32GB MicroSDHC Class10 UHS-1 Speed Up To 45MB/s I gave a link to some posts above:
 

I tested this device to make sure the entire memory existed (not a fake card) and to check the speeds. With my laptop's built in SD card reader I got the following results:

Writing speed: 14.1 MByte/s
Reading speed: 50.8 MByte/s

Class 10 mandates that the minimum read/write speed must be 10MB/s, this card exceeds that specification.

The Actual product being offered and shipped by Amazon is the same model number, but the manufacturer (Transcend Taiwan) specifications are actually different. This is probably why there are a few reviews with people saying the speed is not as fast as expected/advertised. I have attached some photos of the actual product received on (2) separate occasions that confirm the true spec is up to 45MB/s sequential read (20MB/s sequential write). If you compare the photos I supply and the ones on the Amazon web site, you will see the Amazon photo does not show the printed speed rating to the right side of the device packaging. Regardless, the speed is fast enough to handle any phone and/or tablet read task which often peak at 15MB/s or less. [...]

Please notice the data provided fits well with the one you measured, so, no, neither your card, nor the adapter are fakes, IMO.

#20 Wonko the Sane

Wonko the Sane

    The Finder

  • Advanced user
  • 16066 posts
  • Location:The Outside of the Asylum (gate is closed)
  •  
    Italy

Posted 14 June 2014 - 02:11 PM

Den,

you do realize that your quote by S.Booth confirms that there may be the same kind of "half-fakes" similar to the ones documented by pointertovoid, don't you?

I.e. cards that are actually "original" from Transcend (and as such not fake) but with inside them lowerspeed components and re-branded as a higher model (i.e. fake).

 

:duff:

Wonko



#21 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 14 June 2014 - 05:09 PM

Allow me to confuse ourselves more by quoting four reviews from flipkart page where I purchased the crap from:
 

Itz scarily fast. . . . Mind blowing MMC with such a cheap price, Credit goes to flipkart. It works classy & makes video capturing easy with my Windows phone., 45mb is true i saw copying in such a speed, So don't look back. Go for it. . . .

 

Received genuine product and swift in delivery... Flipkart known for, as always...

Was little apprehensive as there were comments that inappropriate product has been shipped and cases of inadvertent delays... But the experience was good as always and is delighted with the product...

Using the card for a month... Lightning speed data transfers and is as fast as internal memory... No drawbacks noted and completely satisfied with the product...

 

First of all, you get almost 30 GB usable space which is acceptable.
And the speed is amazing! I only got to run a few tests as my laptop's card reader was having loose contacts. So i tested by putting it in the phone, and plugging it through the USB, and STILL i got writing speeds of 20 mbps. Satisfied!

File transfers are all super fast! If you want a 32GB card then dont be confused, go for this!

 

The product seem to be average in performance has it show the read and write speed 48 mb/s but it only gives 22 mb/s read and 14 mb/s write speed and a big fault in product package is that their is no sticker about the manufacturing date and price if it get damage how would we claim for the warranty.



#22 Wonko the Sane

Wonko the Sane

    The Finder

  • Advanced user
  • 16066 posts
  • Location:The Outside of the Asylum (gate is closed)
  •  
    Italy

Posted 14 June 2014 - 06:58 PM

Well that is not particularly confusing.

Comparing this:
http://www.flipkart....CCDJCVZ2WTBQQFH
with this:
http://www.flipkart....CCDVP75Z6MGE8MB
may.

At first sight the speed is only on the actual package and the micro SD has identical markings.
http://en.wikipedia....ed_class_rating

The "Class" as often happens when you allow supposed IT experts to create standards together with supposed marketing experts means nothing, both are SD HC and BOTH are Class 10, yet the manufacturer provides for them (in the package) "up to" speeds that are one the DOUBLE of the other.

Any five years old would understand that it makes no sense to "define a class" and invent a logo made out of the speed in Mb/s corresponding to the class inside a "C" (not a particularly "creative" idea), you just make no "class" and tell manufacturers to write the minimum guaranteed speed in Mb/s inside a big C, while at the same time PROHIBITING them from writing ANYWHERE "up to x Mb/s".

Then, since I like "aggressive" standards, the SD association:
http://en.wikipedia..../SD_Association
should buy randomly samples online for EACH make/model, test if the average of speed of (say) ten samples from different sources (along a known, common protocol of testing) reflects the number inside the big C, and if it does not, take back from the manufacturer the attribution of member of the association and the SD markings license for no less than 12 months, and promote via internet a public campaign to let the users know how company x or y has been revoked from membership because of a violation of the standard, producing products marked in a deceiving or downright wrong way.

Companies that wish to have a higher speed rating, will need to guaranteee that minimum speed, not mark the actual item with a lower standard and then put on their site or on the package an "up to" piece of info.

Consider the speed rating of tires:

Would you put on your Porsche 911 a tire marked S, T or U because on the manufacturer site or on the label coming with the tire there is written "up to 320 kmh"?

For all we know the stupid thingy could be made with a process that in (say) 72.78% of cases allows a "top speed" and in (still say) the remaining 27.22% only goes slightly over the "Class 10" speed, the manufacturer marks them as C10, and is allowed to write "up to a zillion fantastibytes/s", then three final users (the lucky ones) get one specimen that can actually reach the advertised speed (and post enthusiastic reviews) and the fourth one gets the slowish one, but cannot protest much, since the marking is "only" C10.

:duff:
Wonko

#23 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 15 June 2014 - 12:27 AM

TS_RDF5.jpg610LFPOsMpL.jpg


Do those images ring any bells? Well, the 1st one is from Holmes.Sherlock's post starting this theread, while the second one is one of the 16 user images in the reviews section of the already cited Transcend 32GB MicroSDHC Class10 UHS-1 Memory Card with Adapter 45 MB-s (TS32GUSDU1E) page at amazon.com, and, there, it has a legend saying: "Kingston MobileLite G3 reader plugged into USB 3.0 port". Please do also bear in mind that, 8 out of 16 pics in the just mentioned set are DiskMark screenshots, and that the numbers in them make no real sense, but, instead, appear to show various different flash media. Which, in fact, they probably do, and Wonko is probably quite right about the "Half-Fakes". But those numbers above show a medium that's noteworthy, in that it attains almost twice the specified (viz. 45 MB/s) max. transfer-rate for sequential- and 512k-random-reads, while still (apparently) bearing the same label (in case it indeed does, it's not obvious which review this image comes from)... Then again, user reviews should always be taken with several grains of salt, however, since the world is full-to-the-brim of clueless people, of course...



#24 Holmes.Sherlock

Holmes.Sherlock

    Gold Member

  • Team Reboot
  • 1444 posts
  • Location:Santa Barbara, California
  •  
    United States

Posted 15 June 2014 - 04:34 AM

But those numbers above show a medium that's noteworthy, in that it attains almost twice the specified (viz. 45 MB/s) max. transfer-rate for sequential- and 512k-random-reads, while still (apparently) bearing the same label (in case it indeed does, it's not obvious which review this image comes from)... 

 

Reads are thrice faster but writes are not. When Transcend or any other company rates a card as 45 MB/s, they do not specify explicitly whether they refer to read or write.



#25 dencorso

dencorso

    Frequent Member

  • Advanced user
  • 142 posts
  •  
    Brazil

Posted 15 June 2014 - 04:54 AM

Though it should refer to sequential read, since it's the one where it's easier to attain good-looking transfer-rates... :D

 

The reads that are thrice faster are those that had maxed-out due to USB 2.0... both tests (the second one is on USB 3.0) should refer to the same card (namely: yours or one one equivalent to it)... in case I'm right, your card should give those great read transfer-rates, when connected to USB 3.0!

 

BTW, I've added info to post #8 and reformulated it, aiming to make it clearer, please do give it another read.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users