Miscellaneous Performance Metrics

This section looks at some of the other commonly used benchmarks representative of the performance of specific real-world applications.

3D Rendering - CINEBENCH

We use CINEBENCH R15 and R23 for 3D rendering evaluation. R15 provides three benchmark modes - OpenGL, single threaded and multi-threaded, while R23 provides only single and multi-threaded modes. Evaluation of different PC configurations in all supported modes provided us the following results.

3D Rendering - CINEBENCH R15 - Single Thread

3D Rendering - CINEBENCH R15 - Multiple Threads

3D Rendering - CINEBENCH R15 - OpenGL

3D Rendering - CINEBENCH R23 - Single Thread

3D Rendering - CINEBENCH R23 - Multiple Threads

Multi-threaded performance is broadly related to the core count, though the DIY configuration with higher TDP budget is also able to put in a good show. The OpenGL (GPU-dependent) performance is obviously in favor of the unit with the Iris Plus GPU. The R23 scores are in line with the clock speeds and core counts of the different CML-U SiPs used in the PCs.

x265 Benchmark

Next up, we have some video encoding benchmarks using x265 v2.8. The appropriate encoder executable is chosen based on the supported CPU features. In the first case, we encode 600 1080p YUV 4:2:0 frames into a 1080p30 HEVC Main-profile compatible video stream at 1 Mbps and record the average number of frames encoded per second.

Video Encoding - x265 - 1080p

Our second test case is 1200 4K YUV 4:2:0 frames getting encoded into a 4Kp60 HEVC Main10-profile video stream at 35 Mbps. The encoding FPS is recorded.

Video Encoding - x265 - 4K 10-bit

The CI662 nano seems a bit hampered by the slow DRAM, but otherwise matches the CI660 nano. Against the DIY Bean Canyon / Akasa Turing kit, the CI662 nano is at a disadvantage because of TDP limitations. The hexa-core Frost Canyon NUC takes the lead in both 8b and 10b encoding due to the large number of threads available to the encoder.


7-Zip is a very effective and efficient compression program, often beating out OpenCL accelerated commercial programs in benchmarks even while using just the CPU power. 7-Zip has a benchmarking program that provides tons of details regarding the underlying CPU's efficiency. In this subsection, we are interested in the compression and decompression rates when utilizing all the available threads for the LZMA algorithm.

7-Zip LZMA Compression Benchmark

7-Zip LZMA Decompression Benchmark

Performance broadly follows the relative core count, clock speed, and DRAM speed ordering for the systems under consideration - and the CI662 nano marks its entry in the middle of the pack.

Cryptography Benchmarks

Cryptography has become an indispensable part of our interaction with computing systems. Almost all modern systems have some sort of hardware-acceleration for making cryptographic operations faster and more power efficient. In this sub-section, we look at two different real-world applications that may make use of this acceleration.

BitLocker is a Windows features that encrypts entire disk volumes. While drives that offer encryption capabilities are dealt with using that feature, most legacy systems and external drives have to use the host system implementation. Windows has no direct benchmark for BitLocker. However, we cooked up a BitLocker operation sequence to determine the adeptness of the system at handling BitLocker operations. We start off with a 2.5GB RAM drive in which a 2GB VHD (virtual hard disk) is created. This VHD is then mounted, and BitLocker is enabled on the volume. Once the BitLocker encryption process gets done, BitLocker is disabled. This triggers a decryption process. The times taken to complete the encryption and decryption are recorded. This process is repeated 25 times, and the average of the last 20 iterations is graphed below.

BitLocker Encryption Benchmark

BitLocker Decryption Benchmark

The CI662 nano, CI660 nano, DIY Bean Canyon NUC / Akasa Turing kit, and the Frost Canyon NUC perform very similarly in this benchmark, given that they are based on processors with the same microarchitecture. It is possible that the updates in Windows 10 20H2 have given the CI662 nano an edge (given that the benchmarks on the other systems were run with either Windows 10 1903 or Windows 10 1909).

Creation of secure archives is best done through the use of AES-256 as the encryption method while password protecting ZIP files. We re-use the benchmark mode of 7-Zip to determine the AES256-CBC encryption and decryption rates using pure software as well as AES-NI. Note that the 7-Zip benchmark uses a 48KB buffer for this purpose.

7-Zip AES256-CBC Encryption Benchmark

7-Zip AES256-CBC Decryption Benchmark

The 7-zip cryptography benchmark numbers are in-line with the results for plain archiving. DRAM speed shouldn't be much of a factor here, as the buffer should easily fit into the cache. 

Yet another cryptography application is secure network communication. OpenSSL can take advantage of the acceleration provided by the host system to make operations faster. It also has a benchmark mode that can use varying buffer sizes. We recorded the processing rate for a 8KB buffer using the hardware-accelerated AES256-CBC-HAC-SHA1 feature.

OpenSSL Encryption Benchmark

OpenSSL Decryption Benchmark

The performance of the different systems for OpenSSL seem to be fairly similar to each other, with the CI662 nano in the middle of the pack.

Agisoft Photoscan

Agisoft PhotoScan is a commercial program that converts 2D images into 3D point maps, meshes and textures. The program designers sent us a command line version in order to evaluate the efficiency of various systems that go under our review scanner. The command line version has two benchmark modes, one using the CPU and the other using both the CPU and GPU (via OpenCL). We present the results from our evaluation using the CPU mode only. The benchmark (v1.3) takes 84 photographs and does four stages of computation:

  • Stage 1: Align Photographs (capable of OpenCL acceleration)
  • Stage 2: Build Point Cloud (capable of OpenCL acceleration)
  • Stage 3: Build Mesh
  • Stage 4: Build Textures

We record the time taken for each stage. Since various elements of the software are single threaded, and others multithreaded, it is interesting to record the effects of CPU generations, speeds, number of cores, and DRAM parameters using this software.

Agisoft PhotoScan Benchmark - Stage 1

Agisoft PhotoScan Benchmark - Stage 2

Agisoft PhotoScan Benchmark - Stage 3

Agisoft PhotoScan Benchmark - Stage 4

The CI662 nano is slightly handicapped by the slower DRAM speed compared to the CI660 nano. The trend in the results are indicative of the faster CPUs with higher core counts performing better.

Dolphin Emulator

Wrapping up our application benchmark numbers is the new Dolphin Emulator (v5) benchmark mode results. 

Dolphin Emulator Benchmark

This is again a test of the CPU capabilities - the CI660 nano and the CI662 nano perform very similar to each other, and the faster Core i7-10710U with two extra cores in the Frost Canyon NUC makes it come out at the top of the pack.

UL Benchmarks - PCMark and 3DMark Storage and Networking Performance


View All Comments

  • wr3zzz - Tuesday, January 5, 2021 - link

    I am a huge fan of passively cooled PC but at this price point there is not much value proposition. I understand that there are no more Y-series CPU at the moment (love my HP Folio) but most notebook at $700 price point today has a silent fan profile if one were to use it as a HTPC. Also, this thing is not a looker either and for aesthetics I would rather have a notebook with display off as HTPC. For everything else its performance is no better than a notebook of equivalent value and I would trade the fan noise for actual utility of having the display and inputs in one unit.

    NUC has no chance if its value proposition is inferior to notebooks.
  • Operandi - Tuesday, January 5, 2021 - link

    Yeah, I have to agree the value of this as a HTPC is absurd which is a shame as lit looks like it works quite well. And from the pics I think it actually looks alright (at least from pics) from ID perspective; much better than the ugly AF Intel NUCs.

    I got a pretty good deal on a older 8000 series NUC and put it in a passive Akasa case. Looks very cool, cools decently but still pretty expensive even if the NUC was way under retail.
  • Hulk - Tuesday, January 5, 2021 - link

    My thoughts exactly. You can buy a laptop for less money. The price point is way off.

    Beside that huge miss it also needs a Tiger Lake processor for better performance and thermals/efficiency.
  • jeremyshaw - Tuesday, January 5, 2021 - link

    I would even argue as a forward looking HTPC, the lack of Tiger Lake kills it. No hardware AV1 decode on these old Skylake+++ CPUs. Reply
  • Jorgp2 - Tuesday, January 5, 2021 - link

    There's always the Atom based NUCs if you just want an HTPC.

    Hopefully the Tremont based NUCs will have 4K 10bit HDR output this time.
  • npz - Tuesday, January 5, 2021 - link

    I have yet to hear a laptop that is actually quiet under load. Who says this mini-PC purpose is for HTPC only? There are much cheaper options for HTPC only PCs.

    You are paying a premium for true silence under total load for actual work e.g. Adobe Creative Suite, etc. that you won't find on any laptop or a fanned NUC/mini-pc. Speaking of which, NUCs are also loud, like all laptops, under load. I personally can't stand my notebook when it's doing encoding or anything video processing related.
  • npz - Tuesday, January 5, 2021 - link

    Custom fitted heatsink pieces like this are just not mass produced:
  • wr3zzz - Tuesday, January 5, 2021 - link

    The Y-series 5W chips are always passively cooled and totally silent but are only for premium products easily over $1000+. Intel just hasn't made one for 10nm yet. The last one was from two years ago.

    Nobody is going to run Adobe Creative Suite or doing any content creation on this machine for that matter. Besides HTPC it's just productivity softwares which you can certainly run silently in no fan mode in recent notebooks if the the design is decent.
  • npz - Tuesday, January 5, 2021 - link

    The Y chips are also correspondingly much weaker, and those are used in the lower end Surfaces if I recall. The higher end laptops/tablets don't use Y series chips

    And why not run Adobe software on a machine like this, with laptop class latest gen i7 cpu?! Even Intel advertises their i7 NUCs with explicit Adobe Premiere editing in their marketing.
    This isn't a low end Celeron or Pentium or Bay Trail class cpu that would be HTPC . Even productivity sotware / office suites can put heavy load on the cpu. Try some heavy spreadsheet calculation for example. PDF software like for OCR would also ramp up fans. Even if you're not editing movies, just editing photos can put 100% load on all cores. Or are you saying no one should edit photos on these machines?
  • npz - Tuesday, January 5, 2021 - link

    And if I can use Adobe CC suite and Handbrake and others on my old Broadwell laptop, I would certainly do the same on this type of machine. Think about movie editing apps in Google Play that's running on much weaker Android devices Reply

Log in

Don't have an account? Sign up now