Putting the Raspberry Pi Zero vs MangoPi MQ Pro was something I’d wanted to do since seeing the MQ Pro’s announcement and specifications. It just seemed to make sense. On paper, they’re largely similar, with 1GHz single-core CPUs, and 512MB of RAM. A 1GB MQ Pro is also available and is what I’ll be using here so your mileage may vary slightly if you have the 512MB version.
Side note: If you’re interested in my content and really bad puns, follow me on Twitter – it’d really mean a lot! Now, back to the data..
Table of Contents
Specifications
| MangoPi MQ Pro | Raspberry Pi Zero W | |
|---|---|---|
| CPU | 1GHz Allwinner D1 C906 RISC-V Processing Unit | 1GHz ARM1176JZF-S – BCM2835 |
| HiFi4 DSP | ||
| RAM | 512/1024MB DDR3L RAM | 512MB LPDDR2 SDRAM |
| GPU | G2D 2D Graphics Accelerator | VideoCore IV |
| 4K capable H.264/H.265 Processing Unit | H.264 MPEG-4 decode (1080p30) | |
| 4Kp30/1080p60 HDMI Output* | ||
| Connectivity | 802.11 b/g/n WLAN (2.4GHz – RTL8723DS) | 802.11 b/g/n WLAN (2.4GHz) |
| Bluetooth 4.2 | Bluetooth 4.1 / BLE | |
| u.FL Antenna Connector | ||
| USB-C 2.0 OTG (1) | Micro USB 2.0 OTG (1) | |
| Mini HDMI (1) | Mini HDMI (1) | |
| 24 pin DVP/RMII Camera Connector (1) | CSI-2 Camera Connector (1) | |
| microSD Card Slot | microSD Card Slot | |
| 40 Pin Raspberry Pi compatible GPIO header | 40 Pin Raspberry Pi GPIO header | |
| Audio via audio out pads | ||
| Power | USB-C (5V) | Micro USB (5V) |
| via GPIO Header | via GPIO Header | |
| Dimensions | 65x30mm (Length x Height) |
Benchmarking Information / Hardware
For these benchmarks, the boot storage is the best-performing Amazon Basics 64GB microSD card and the SSD is a Samsung 850 EVO (500GB). For USB Ethernet, I use the Linksys 1Gbit USB Ethernet adapter. A 120mm fan is blowing across the board for all tests (unless mentioned) to ensure no thermal throttling takes place.
Raspberry Pi Zero vs MangoPi MQ Pro – Compute Performance
As always, I’ll start with UnixBench and move on to a mix of “real-world” and other synthetic benchmarks.
The Mango Pi MQ Pro under PHPBench testing shows a 50.31% performance increase over the Raspberry Pi Zero W and its compute dominance doesn’t end there. PyBench also completes 22.26% faster and the WavPack encoding test is 7.43% quicker on the MangoPi board.
When we move to GZIP compression, Crypto++ and OpenSSL work though, the Pi Zero W starts to fight back. With GZIP, it’s 34.98% faster to complete the same test, OpenSSL benchmarks show around 50% increases across the board and whilst with Crypto++’s Integer and Elliptical curve run the MQ Pro pulls ahead by 31.23%, Keyed and Unkeyed Algorithm runs were 20.73% and 10.91% faster respectively.
UnixBench
PHPBench
PyBench
WavPack Audio Encoding
Crypto++
OpenSSL Bench
GZIP Compression
Memory (RAM) Performance
It should be noted that we are testing the 1GB RAM model of the MangoPi MQ Pro here. If you compare it to a 512MB board, your results will likely differ! The MQ Pro also gets an advantage from its newer DDR3 memory and this is clearly shown across the board with tinymembench and CacheBench with performance increases between 20% and nearly 700%!
tinymembench
CacheBench
Networking Performance
The MQ Pro train doesn’t stop when we move onto networking performance either, with WiFi tests showing it 60% and 17.37% faster during download and upload testing. We see a slightly odd result with a USB Ethernet adapter where the Pi Zero offers a 16.66% gain on the download side, though trails by 26.27% on the upload front.
Storage Performance
MicroSD Card Reader
The micro SD card results are taken from my Best SD Cards for Raspberry Pi piece and show that in comparison to other similar boards, the SD card reader in the MQ Pro isn’t exactly the speediest of things. With the board’s limited compute resources though, this shouldn’t hold it back too much unless you’re hoping to run a speedy full desktop environment but top tip, don’t.
MangoPi MQ Pro
| SD Card Model | Sequential Read | Sequential Write | Random Read | Random Write | IOPing |
|---|---|---|---|---|---|
| SanDisk Ultra (8GB) | 11.4 MB/s | 8.57 MB/s | 4.88 MB/s | 0.61 MB/s | 1 ms |
| SanDisk Ultra (16GB) | 11.4 MB/s | 9.07 MB/s | 4.56 MB/s | 2.07 MB/s | 2.95 ms |
| SanDisk Ultra (32GB) | 11.42 MB/s | 9.37 MB/s | 4.6 MB/s | 1.87 MB/s | 3 ms |
| SanDisk Extreme (64GB) | 11.42 MB/s | 10.8 MB/s | 3.18 MB/s | 2.68 MB/s | 0.96 ms |
| SanDisk Extreme PRO (128GB) | 11.4 MB/s | 10.8 MB/s | 2.74 MB/s | 2.63 MB/s | 0.99 ms |
| Kingston Canvas Select Plus (32GB) | 5.73 MB/s | 5.57 MB/s | 2.97 MB/s | 1.51 MB/s | 1.19 ms |
| KIOXIA EXCERIA (32GB) | 11.42 MB/s | 9.67 MB/s | 5.27 MB/s | 3.08 MB/s | 0.93 ms |
| Samsung EVO Plus (32GB) | 11.39 MB/s | 10.63 MB/s | 4.3 MB/s | 1.55 MB/s | 1.45 ms |
| Amazon Basics (64GB) | 11.48 MB/s | 10.77 MB/s | 5.51 MB/s | 3.25 MB/s | 1.18 ms |
| Verbatim Premium (16GB) | 11.41 MB/s | 10.3 MB/s | 4.9 MB/s | 1.83 MB/s | 1.08 ms |
| SanDisk MAX ENDURANCE (32GB) | 11.42 MB/s | 10.7 MB/s | 3.1 MB/s | 2.47 MB/s | 0.87 ms |
| Integral ultima PRO (64GB) | 11.48 MB/s | 10.8 MB/s | 5.12 MB/s | 3.03 MB/s | 1.42 ms |
| Patriot EP Series (64GB) | 11.41 MB/s | 8.03 MB/s | 4.46 MB/s | 2.13 MB/s | 0.98 ms |
| Kodak (64GB) | 11.38 MB/s | 10.6 MB/s | 2.68 MB/s | 0.75 MB/s | 1.05 ms |
| Intenso (64GB) | 11.47 MB/s | 10.8 MB/s | 5.46 MB/s | 3.03 MB/s | 1.23 ms |
| Transcend (32GB) | 11.48 MB/s | 8.03 MB/s | 5.12 MB/s | 2.06 MB/s | 0.86 ms |
| Samsung EVO Select (32GB) | 11.38 MB/s | 10.6 MB/s | 4.49 MB/s | 1.45 MB/s | 1.45 ms |
| Samsung PRO Plus (128GB) | 11.48 MB/s | 10.47 MB/s | 4.61 MB/s | 1.13 MB/s | 0.96 ms |
| SanDisk HIGH ENDURANCE (64GB) | 11.4 MB/s | 10.57 MB/s | 2.85 MB/s | 2.46 MB/s | 0.99 ms |
| Samsung PRO Endurance (32GB) | 11.38 MB/s | 10.63 MB/s | 3.99 MB/s | 1.42 MB/s | 1.33 ms |
Raspberry Pi Zero W
| SD Card Model | Sequential Read | Sequential Write | Random Read | Random Write | IOPing |
|---|---|---|---|---|---|
| SanDisk Ultra (8GB) | 21.11 MB/s | 13.53 MB/s | 4.83 MB/s | 0.66 MB/s | 1.21 ms |
| SanDisk Ultra (16GB) | 21.19 MB/s | 12.77 MB/s | 4.9 MB/s | 2.01 MB/s | 3.04 ms |
| SanDisk Ultra (32GB) | 21.23 MB/s | 15.8 MB/s | 4.87 MB/s | 1.82 MB/s | 3.05 ms |
| SanDisk Extreme (64GB) | 21.25 MB/s | 19.07 MB/s | 4.53 MB/s | 2.68 MB/s | 1.15 ms |
| SanDisk Extreme PRO (128GB) | 21.23 MB/s | 18.57 MB/s | 4.43 MB/s | 2.7 MB/s | 1.18 ms |
| Kingston Canvas Select Plus (32GB) | 11.03 MB/s | 9.8 MB/s | 3.17 MB/s | 1.68 MB/s | 1.24 ms |
| KIOXIA EXCERIA (32GB) | 21.21 MB/s | 14.23 MB/s | 5.1 MB/s | 3.04 MB/s | 1.14 ms |
| Samsung EVO Plus (32GB) | 20.98 MB/s | 17.03 MB/s | 4.68 MB/s | 1.79 MB/s | 1.29 ms |
| Amazon Basics (64GB) | 21.36 MB/s | 19.6 MB/s | 5.49 MB/s | 2.98 MB/s | 1.38 ms |
| PNY Performance Plus (16GB) | 21.15 MB/s | 10.8 MB/s | 4.7 MB/s | 2.14 MB/s | 1.36 ms |
| Verbatim Premium (16GB) | 20.92 MB/s | 10.4 MB/s | 3.02 MB/s | 0.75 MB/s | 1.27 ms |
| SanDisk MAX ENDURANCE (32GB) | 21.18 MB/s | 18.9 MB/s | 4.42 MB/s | 2.46 MB/s | 1.06 ms |
| Integral ultima PRO (64GB) | 21.29 MB/s | 19.5 MB/s | 4.88 MB/s | 2.9 MB/s | 1.49 ms |
| Patriot EP Series (64GB) | 21.18 MB/s | 17.37 MB/s | 4.39 MB/s | 2.22 MB/s | 1.15 ms |
| Kodak (64GB) | 21.02 MB/s | 17.97 MB/s | 2.56 MB/s | 0.77 MB/s | 1.24 ms |
| Intenso (64GB) | 21.33 MB/s | 19.63 MB/s | 5.52 MB/s | 2.92 MB/s | 2.25 ms |
| Transcend (32GB) | 21.15 MB/s | 12.27 MB/s | 5.02 MB/s | 2.08 MB/s | 1.06 ms |
| Samsung EVO Select (32GB) | 20.91 MB/s | 17.23 MB/s | 4.82 MB/s | 1.64 MB/s | 1.33 ms |
| Samsung PRO Plus (128GB) | 21.14 MB/s | 17.17 MB/s | 4.4 MB/s | 1.19 MB/s | 1.21 ms |
| SanDisk HIGH ENDURANCE (64GB) | 21.2 MB/s | 18.6 MB/s | 3.75 MB/s | 2.42 MB/s | 1.17 ms |
| Samsung PRO Endurance (32GB) | 20.81 MB/s | 16.93 MB/s | 4.89 MB/s | 1.62 MB/s | 1.28 ms |
USB SSD & microSD Card Comparison
The SD data here is an average of all tested microSD cards so that there’s a more simple comparison.
The USB SSD testing surprised me though. Bar the odd sequential read results they’re much the same, with the BCM2835 based Raspberry Pi being slightly ahead. I tried with 2 different USB to SATA adapters and 2 SSDs to try and find what was going on with the sequential read results and each was consistently that far ahead! The only consistent difference was that I used a USB-A to Micro USB adapter on the Pi Zero and USB-A to USB-C on the MQ Pro but I don’t believe this to be relevant given the other tests falling in line.
Thermals & Power
Temperatures
With an ambient room temperature of 25.8 degrees celsius in both tests we can see that these 2 boards are quite close in terms of operating temperatures. What interested me during this testing was the delta between idle and load on the MQ Pro. As I mentioned in my stand-alone MangoPi MQ Pro Review, I thought I may have been doing something wrong, or stress-ng wasn’t up to the task but with multiple forms of stress-testing I saw the same behaviour.
Power Draw
Finally, we move on to the overall power draw where the Pi Zero W sips 17.74% more power at idle and a whopping 34.48% more under load. Whilst you may look at the numbers and see 0.73W vs 1.17W and think that’s not a big difference, if you’re thinking of running either of these boards on battery or solar power, this will make a big difference to your potential run-time.
Closing Thoughts
There’s a theme running throughout these tests and it’s one of surprise for me. I’ll be very honest and say I knew very little about the RISC-V ecosystem before diving in with the MangoPi MQ Pro but I’m so glad that I did. The process of getting it up and running and seeing how it compared to its closest “rival” in the Zero style board lineup was great.
The issue at this point though is that due to MangoPi being a smaller, newer manufacturer with limited resources, they’re not churning out as many units and due to the initial buzz, they’re being snapped up quickly. This means that whilst yes, the MQ Pro is a very valid alternative to the Pi Zero if your software/interests will run on a RISC-V architecture, annoyingly, you still need to wait to get your hands on one.
That also ignores the price side of things. The Raspberry Pi Zero W retails for around £10GBP (keep an eye out on rpilocator if you’re currently in the market) in the UK through authorised retailers whereas the Mango Pi MQ Pro 1GB model tested here will run you around £23 if you manage to get one through the official store when they have stock (these prices both include GST/VAT at 20%). At £23 I still think it’s worth it to get your hands on a small RISC-V based board that offers twice as much, faster RAM and better performance in a lot of areas but if you’re purely interested in the price this may not appeal to you.
Good job, mangobuge and anyone else behinds the scenes! You’ve created a great product with huge potential. With the ARM H616 MangoPi MQ Quad on the way, it’s great to have more alternatives available. On behalf of myself and the community though, please make sure that the operating system images and support is available upon future board launches!
7 comments
I would have rather seen the 512’s benchmarking rather than Mango Pi’s 1Ghz tested. Just from these numbers, it looks like the Zero 512 would be way better than the Mango’s 512. But the author of the article was probably paid to do this test by the Mango’s maker. Raspberry Pi foundation would have paid for the comparison that I want to see.
When you say 1GHz did you mean 1GB in regards to the RAM on the Mango Pi MQ Pro board?
If I was paid to write this I would have mentioned that it was a paid piece clearly at the beginning but it was not so I didn’t and I’m not too sure what you’re trying to achieve by saying I was? I’m not a professional reviewer/tester, I do this in my spare time out of pure curiosity and interest and as I purchased the 1GB MQ Pro that’s what I used. I clearly state that multiple times throughout and note that the performance may differ if you have a 512MB model because of it. I wish I was getting paid for this but sadly not, I even choose to not run advertising on the site so 🤷♂️
In the interest of transparency, I should update this to say that I am now experimenting with display advertising as a way to try and help fund the purchase of further boards and accessories. If after a week or so it doesn’t make sense and only creates a negative experience for users then I’ll remove them.
This is a very interesting comparison, and I fully agree to the conclusion that it’s always good to have more options to choose from.
But the results for the MQ Pro look rather disappointing considering it (and it’s SOC) are brand new models. It barely keeps up with an ARM chip from 2003. And for 2022 vs. 2003 the outcome should be in a whole other ball park. It seems there’s currently a lot of fluff around RISC V, but it still has a very long way to go until it’s competitive against the established architectures.
Also I wouldn’t agree the MQ Pro is generally a good offer. It sure is, if you require the really tiny form factor and don’t care about the ecosystem and available software. If you are okay with just a little larger, there are plenty of boards with more recent ARM systems that offer multiple cores of which every single one is faster than the D1.
So it’s well worth watching how RISC V develops. Comparisons like this one are essential for this and hence very welcome! Still RISC V doesn’t seem to be ready for the majority of users right now.
Absolutely. It will be interesting to see how things move on and how the quad-core Mango Pi board looks.
In regards to it being a good offer, I’m not sure if I implied that as such, apologies if that’s how it came across :D I just wanted to say that at the official store’s price point, it’s not too high of a price if you’re wanting to play with a RISC-V-based SBC. If you can get your hands on a Raspberry Pi Zero 2 W at RRP then go for that if you want a faster, better-supported board for now!
Hey Bret, thanks for the info.
I’m so happy to try new risc arch!
Have you tried to set up v4l2 on this board? I can’t make it work so far, I wonder if cedrus supports this board at all. :)
Thanks!
Hi, Dmitry! It’s not something I’ve tried myself I’m afraid :( If you get it working, come back and let me know how!