## Comparing Raspberry Pi to Apple II

Posts: 3
Joined: Thu Apr 11, 2013 9:27 am

### Comparing Raspberry Pi to Apple II

Hi All
I'm working on an infographic to send out to schools about the Raspberry Pi and one element of it is a visualisation of the affordability of the Pi's computing power compared to the Apple II of 1977. I am however a filmmaker (interested in the creation of 'intelligent' cameras) with not much computer science knowledge and have got myself hopelessly out of my depth with the benchmarking calculations.

I've put this on my bloghttp://www.bristolfilmacademy.com/bfa-b ... pples.html, where I've had one response, but thought maybe someone here had already tried to do a similar thing and could help me...

The figure I've arrived at is 93 million times more affordable - but my methods *may* not have been rigorous enough to go to print with that figure. Here's how I arrived at the number (please don't laugh).

I found this benchmarking information, http://www.roylongbottom.org.uk/whetsto ... nchorAcorn which was the only way I could find a common currency between the Apple II and the processors of today: the MWIPS, or Millions of Whetstone Instructions per Second. The Apple II clocked 0.0011 MWIPS.

Unfortunately, there's no benchmarking for the Raspberry Pi, nor any ARM chips, so the best I could do was take the Core 2 Duo M speed - 1,557 MWIPS - as an estimate, based on the fact that I heard a bloke in the pub say that the Pi is equivalent to the computers of five years ago.

So, I'd calculated that the Raspberry Pi is something like 1.415 million times faster than the Apple II, but this is only half the story.

When the Apple II was launched, it was also available as a 'board only' option, which seems like the best option for comparison with the Raspberry Pi. As you can see, this cost \$798.

My next step was to adjust for inflation using this calculator, http://www.westegg.com/inflation/ and that came out at \$2,981.

The Pi costs \$45, so is 66 times more affordable.

So, I make Raspberry Pi's computing power roughly 66 * 1.415m = 93 million times more affordable than that of the Apple II.

(As the comment on my blog pointed out, it's more complicated than that. The ARM is slower but overall the performance is faster because of the GPU, but how would you factor that in).

Any help greatly appreciated.

Matt

Davespice
Forum Moderator
Posts: 1662
Joined: Fri Oct 14, 2011 8:06 pm
Location: The Netherlands

### Re: Comparing Raspberry Pi to Apple II

Hmm, I think one of the hardware guys might be able to give you a more accurate figure on the MWIPS the Pi is capable of. Gert / Dom maybe?

I think the best way to represent this would be pence/cent per MWIP
Show this value for both machines, with the inflation adjusted value for the Apple
And only then show the calculation of affordability.

I like the idea though, would be interested to see it when its finished

Posts: 3
Joined: Thu Apr 11, 2013 9:27 am

### Re: Comparing Raspberry Pi to Apple II

Thanks Dave

Yes, that would be a nice way of doing it.

The response on the blog suggested that the MWIPS of the ARM chip was much less than the Core Duo at 30 - 60 MWIPS but pointed out that it wasn't a true comparison because of the GPU, which he thought increased the effective computing speed by one or two orders of magnitude.

I don't know whether there'd be an acceptable way of converting the effect of the GPU into MWIPS....

rurwin
Forum Moderator
Posts: 4258
Joined: Mon Jan 09, 2012 3:16 pm
Contact: Website

### Re: Comparing Raspberry Pi to Apple II

The Whetstone test measures floating-point performance, which the Apple ][ was unfairly bad at, and which forms a minor part of the workload of any computer. You would be on firmer ground comparing Dhrystones.

Both figures for the Raspberry Pi are here. The Drhystone rating for the Apple ][e was 0.021MIPS, which needs to be multiplied by 1757 to get Dhrystones per second: 36.897

The RaspPi does 809,061.5 Dhrystones per second, so that's a difference of x21,928
The same page gives the RaspPi's whetstone rating as 667 MIPS, giving a factor of x606,091. The numbers are so close that if you adjusted for the mix of floating-point and integer work, (at least 5:95) the number would be close enough to the integer figure that it wouldn't be worth doing. (What's a factor of 2 here or there? )

Any factor you introduce for the GPU is going to be wrong, maybe by orders of magnitude, since it depends critically on what the GPU is being asked to do.

If we assume that one Drhystone instruction can write one pixel, then an Apple ][ could write 0.021 megapixels per second, and the RaspPi can write 1 gigapixels per second. That's a factor of 47,619, but it doesn't account for all the GPU hardware.

So depending on the amount of graphics, floating-point and integer work, the answer is somewhere in the range 20,000 (all integer, no graphics) to 100,000 (videocore and ARM fully occupied with 5% floating-point). An unrealistic maximum would be 650,000 for full graphics and 100% floating point. You might as well make that a million to account for the other GPU hardware. My feeling is that 50,000 would be a reasonably conservative figure that could be asserted with some authority.

Davespice
Forum Moderator
Posts: 1662
Joined: Fri Oct 14, 2011 8:06 pm
Location: The Netherlands

### Re: Comparing Raspberry Pi to Apple II

It is hard to make a good, scientifically water tight, comparison since the GPU is a form of parallel processing. However you could go in the opposite direction and make a less scientific comparison but one that your target audience might understand more.

For example, regardless of how the computer is implementing the task, compare the time required to perform the same task on both machines.

If you remember Micro Live from the 80’s they compare an Archimedes to a BBC Master by getting them both to draw the same graphical algorithm.

Posts: 3
Joined: Thu Apr 11, 2013 9:27 am

### Re: Comparing Raspberry Pi to Apple II

Yes, that definitely sounds like a more meaningful comparison. Anyone got an Apple II?

RoyLongbottom
Posts: 281
Joined: Fri Apr 12, 2013 9:27 am
Location: Essex, UK
Contact: Website

### Re: Comparing Raspberry Pi to Apple II

You might also be interested in the following.

Whetstone, Dhrystone and other benchmarks on ARM CPUs with results. Source code downloads are provided. (all FREE with no Ads).

http://www.roylongbottom.org.uk/android ... hmarks.htm

Old Dhrystone results including Apple IIe

http://performance.netlib.org/performan ... .col0.html

Results on PCs

http://www.roylongbottom.org.uk/dhrystone%20results.htm
http://www.roylongbottom.org.uk/whetstone%20results.htm

Roy

RoyLongbottom
Posts: 281
Joined: Fri Apr 12, 2013 9:27 am
Location: Essex, UK
Contact: Website

### Re: Comparing Raspberry Pi to Apple II

I forgot my benchmarks on Linux based PCs. Details and results are in the following and it includes links to source codes used to compile the programs for Linux, in tar.gz files.

http://www.roylongbottom.org.uk/linux%20benchmarks.htm

Roy

geev03
Posts: 84
Joined: Thu Jun 07, 2012 12:40 pm
Location: London, UK

### Re: Comparing Raspberry Pi to Apple II

" The RPI2 and 3 are tens of thousands times faster than the Apple IIe. I’ll try to expand on that.."https://www.quora.com/How-much-more-pow ... -Apple-IIe

jamesh
Raspberry Pi Engineer & Forum Moderator
Posts: 23078
Joined: Sat Jul 30, 2011 7:41 pm

### Re: Comparing Raspberry Pi to Apple II

Winner! Oldest thread necro of the month!

2013->2019. 6.5 years!
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."