Note: Versions R9a-x have now been updated to R9z.
We are pleased to release R9 (version 31.66) update for Sandra 20/21 with the following changes:
Happy New Year 2022!
After a good test, we are now shipping Sandra with ARM64 support for Windows – running on compatible ARM64 devices from Arm (Cortex v8 AArch64), Qualcomm (e.g. Snapdragon / Kryo), Broadcom (e.g. Raspberry Pi 3B, 3B+, 4B). Sandra supports all modern ARM64 features like NEON (Advanced SIMD), Crypto hardware acceleration (AES, SHA1, SHA2), VFP, etc. [Note: currently SVE is not supported by Microsoft’s compiler / Windows, and will be included in a future update]
Sandra was perhaps of the first mainstream benchmark to originally natively support 64-bit Windows – first with IA64 (Intel Itanium) then with x64/EM64T (AMD Opteron 64) and finally ARM v7 (32-bit AArch32) support. At CES 2022 we have finally seen many more cool ARM64 Windows devices coming out to battle AMD and Intel. All we need is some servers as well. Is this the end for x86?
If you have a Raspberry Pi around – as an computer enthusiast, you are bound to have at least one – maybe now is the time to try Windows 11 or 10 on it? Prepare to overclock it, get some decent cooling and tweak the storage for decent performance. With Sandra you can now check how native ARM64, old ARM32 as well as x86 code performs on Windows ARM64.
We are continuing to make multi-threading improvements across all benchmarks and other modules, including hybrid (big/LITTLE, DynamiQ) architectures, both x86/x64 and ARM/ARM64. Further scheduling and benchmark optimisations are ongoing.
We have a brand-new dynamic workload scheduler for the memory and cache bandwidth benchmarks which takes into account each core’s performance – similar to the processor benchmarks workload scheduler. This way the bandwidth of different cores (big & LITTLE) can be measured in parallel. This works both on the standard x86/x64 platform as well as traditional ARM big/LITTLE platforms.
Please don’t forget to submit benchmark results to the Official SiSoftware Ranker! Many thanks for your continued support.
And please, don’t forget small ISVs like ourselves in these very challenging times. Please buy a copy of Sandra if you find our software useful. Your custom means everything to us!
Hardware Support, Updates & Fixes
- ARM64 (AArch64) Support
- ARM Cortex AArch64 CPU/SoCs including A7X, A5X, X1
- Qualcomm Snapdragon 7c / Kryo CPU/SoC [in R9z]
- Apple M1 (through virtualization, e.g. Parallels VM) [in R9z]
- Microsoft SQ1, SQ2 (Surface X) [in R9z]
- Broadcomm BCM CPU/SoC in Raspberry Pi 3B*, 3B+*, 4B
- Note: BCM SoCs do not include AES/SHA/SM hardware acceleration (!!!)
- Hardware
- Intel “RaptorLake” (RPL) 13th Gen Core preliminary support (S/H)
- Resolved memory reporting issues on Intel “AlderLake” (ADL) 12th Gen Core, “RaptorLake” (RPL) 13th Gen
- ARM NEON support for both AArch64 & AArch32
- Note: BCM SoCs nor any current Windows device support SVE (Scalable Vector eXtensions).
- Memory & Cache Benchmarks
- Dynamic bandwidth workload allocator for big/LITTLE arch (e.g. ARM DynamiQ, Intel Hybrid)
- Buffered Memory Bandwidth
- Direct Cache L1D, L2, L3, L4 Bandwidth
- Processor Cluster Support
- Two (2) cluster support for Intel Hybrid [x64]
- Three (3) cluster support for AMD DynamiQ [Arm64]
- Dynamic bandwidth workload allocator for big/LITTLE arch (e.g. ARM DynamiQ, Intel Hybrid)
- Inter-Thread/Core/Module/Node/Package Transfers
- Inter-big/P-Cores Only Latency additional tests
- Inter-LITTLE/E-Cores Only Latency additional tests
- Cache Latency
- big / LITTLE core latency test (L1D/L1I P/E, L2 P/E, L3, L4)
- Memory & Cache Bandwidth
- big / LITTLE cores bandwidth (L1D/L1I P/E, L2 P/E, L3, L4)
- New Operating System Support:
- Windows 11 (x86, x64, ARM, Arm64)
- Windows Server 2022 LTSC (21H2) (x64)
- Windows Server vNext LTSC (2025*) (x64, Arm64)
Reviews using Sandra 20/21:
- SiSoftware
- Microsoft / Surface Pro X
- Broadcom SoC / Raspberry Pi
- TPMs
- Intel CPU / SoC
- Intel 12th Gen Core AlderLake (i9-12900K(F)) Review & Benchmarks – big/LITTLE Performance
- Mythical Intel 12th Gen Core AlderLake 10C/20T big/P Cores (i11-12999X) – AVX512 Extrapolated Performance 😉
- Mythical Intel 12th Gen Core AlderLake 40C/40T E-Cores (i11-12911E) – AVX2 Extrapolated Performance 😉
- big/Performance Core Performance Analysis (Intel 12th Gen Core AlderLake (i9-12900K))
- Microsoft / Surface Pro X
- SiSoftware Official Ranker
- Intel Arc “Alchemist” A380 Graphics
- Microsoft SQ2 (Microsoft Surface X)
- Raspberry Pi 4 Model B (Broadcom BCM2711)
- Raspberry Pi 3 Model B+ (Broadcom BCM2837)
- Apple Silicon on Parallels VM
Update & Download
Commercial version customers can download the free updates from their software distributor; Lite users please download from your favourite download site.
