The iFlash used with a CF card as shown in my previous article will double the runtime compared to the original Hard Drive, but how does the CF compare to using my latest gen SD-CF Adapter and an SD card?
So for this test I decided to use the newest 7g iPod Classic, as this was donated to me for test. As SD card technology uses a serial transmission method, I expect power consumption to increase with increasing card speeds. So I opted to test a slower SDHC card and a new UHS-I SDXC in my comparison with the CF card.
- 7g iPod Classic
- New 850mAh Battery (thick iPod battery)
- 5th Gen SD-CF Adapter
- 266x 32Gb Compact Flash card
- Class 10 8Gb SDHC card
- UHS-I 64Gb SDXC card
All the cards have exactly the same files added in the same order after the iPod is restored using iTunes.
Power Consumption First
So before we do runtime tests, it would be good to get some baseline power consumption measurements of the various combinations. So after connecting the iPod to my bench power supply, the probes of my current meter and logger are attached and the measurements can begin…
As we have seen the peak current drawn varies by a large amount between the test cards – you pay quite a heavy current draw penalty by using the ultra-fast UHS-I card. However, as can be seen by the average current draw over 60 minutes – and the final average power consumption figure, the difference is minimal between the cards.
This can be explained by the fact that while the faster card draws higher peak currents, it probably spends a lot less time doing so. The access times for the slower cards will much longer than the faster card – so while the slower cards may draw lower peak currents, they will draw it for much longer.
So to see that visually I setup my digital storage scope to measure the active time for the drive i/o circuitry on the iPod – this will show us how long the iPod spends actually reading from the ‘drive’. I will also measure the time required to boot from reset, and how long the iPod takes to fill its read-ahead cache with each card.
The screenshots below shows the drive i/o activity playing a video file. The CF card takes 7.71s to fill the read-ahead cache, while the faster SDXC card 5.23s, so the iPod spends close to 50% longer accessing the drive when using the slower CF card.
So the data implies that the runtimes maybe the same between CF and SD, but one thing the data above does not take in to account – Internal Resistance. Internal Resistance (IR) for the layman is the batteries ability to deliver power.
Quick example, if we have a perfect 10v battery which has 0ohm IR and it is powering a 10watt load – the voltage at the load will be 10v and the current flowing from the battery will be 1Amp. Now if our perfect battery is not so perfect and and actual IR is 2ohm, in our experiment we will find that the voltage at the load is nearer 8.25volts and the current flowing is about 0.825amps givings us power delivery of just under 7watts.
You’re probably wondering why the lecture on batteries, but what happens with the majority of battery technologies is that as the battery discharges the IR increases, and dramatically – so for Li-Ion as used in the iPod the IR can easily double or triple when the battery is nearly empty.
The iPod will have a switch off point to protect the battery, I would think it is around the 3v-3.2v point. Near that voltage the li-Ion is pretty much discharged and the IR is very high. So it is possible that the higher current draw from the SDXC card compared to the CF is enough to cause the battery terminal voltage to sag below the iPods switch point. The result of which will show the CF card as running much longer – we will see if that is the case….
Battery Run Time Tests
From full charged to switch off how long will the iPod operate using each card. Method is simple, full charge the battery – reset the iPod and hit play, time how long it works for!!!
The graph shows the runtime for Music and Video – In a sense from best case to worst case, so in everyday use your own personal runtime should rest somewhere between maximum video runtime to the maximum music playback.
Compared to the CF in music playback the SDXC suffers a 20% decrease, and the SDHC a 10% decrease in runtime. For Video, the CF is about the same as the SDHC but the SDXC see a 5% decrease in runtime.
So for the average user who probably spends about 30% of the time fast forwarding, queuing music, and changing volume (i.e. backlight on), and 70% backlight off listening to the music. I suspect compared to the CF, a SDHC user will see about 5% reductions in total runtime. While a SDXC user will probably see about 10% runtime reduction.