Difference between "free" binaries available on Github and purchasing a "license"

[Obaid Khalid]

Hello,

I wanted to inquire about the difference between purchasing a Doubango ALPR SDK license and using the available libraries on GitHub.

1. Do we get additional personalized support on purchasing a license? 

2. What services are offered for the "yearly commercial support" license?

Obaid

[Mamadou DIOP]

Hi,

Based on your previous messages with the support team you’re looking to purchase mobile licenses and interested in Jetson with JetPack 4.6:

- for mobile licences there is no personalised support, we offer best effort support via this group. Quote from https://www.doubango.org/pricing.html: “https://www.doubango.org/pricing.html”

- there is no "yearly commercial support”. Yearly support is proposed to companies with at least 1,000 mobile licenses and it includes technical support and assistance

- most importantly, we’ll *not* add support for JetPack 4.6 even if you purchase licenses. We already support JetPack 5.1 and later

--
You received this message because you are subscribed to the Google Groups "doubango-ai" group.
To unsubscribe from this group and stop receiving emails from it, send an email to doubango-ai...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/doubango-ai/d80326f9-c367-4279-bfa3-e2b071c56f7bn%40googlegroups.com.

[Obaid Khalid]

Hello,

Yes, we are looking to purchase at least 1000-1500 mobile licenses/year and are interested in JP4.6. 

I also understand that you are already supporting JP 5.1 or later, however, since JP 5.1 is only compatible with Jetson Xavier and above, it does not fulfill our requirement of using Jetson TX2.

Since we plan to purchase at least 1000-1500 licenses per year, we also will want to purchase the "yearly commercial support". I would greatly appreciate it if we could schedule a Teams/Zoom or a phone call to further discuss this issue.

Thank you.

[Mamadou DIOP]

Officially we support 4.4.1 and 5.1.0 but I have seen the SDK run on different versions. Have you tried on 4.6 and got issues? If yes, please describe the issue and share the logs.

To view this discussion on the web visit https://groups.google.com/d/msgid/doubango-ai/f7ac6323-8e8d-4429-9838-44c1373074c3n%40googlegroups.com.

[Martin Lechner]

Hello,

let me step in here for my colleague to give you details about our issue.
Yes, we have tried it on JetPack 4.6.2, and the main issue is that Nvidia went from TensorRT 7 (7.1.3) in 4.4.1 to TensorRT 8 (8.2.1) in 4.6.2. As you already said, the SKD will run on other versions of Jetpack when at least the major TensorRT and CUDA match. Just for reference, we also tried running the 5.1.0 implementation without success.

Some more details:
On Jetpack 4.6.2, the compilation will fail with the message
message: [CompVSharedLib] Failed to load library with path=/home/nvidia/axess_eval/ultimateALPR-SDK-master/binaries/jetson/aarch64/libultimatePluginTensorRT.so, Error: libnvinfer.so.7: cannot open shared object file: No such file or directory
indicating it cannot find the libraries for TensorRT 7. We also tried adding symlinks, but this just led to a crash in the Nvidia libraries - which is expected due to the different major releases. When trying the 5.1.0 binaries, which are built against TensorRT 8 (but a different minor version), we get a mismatch in the CUDA version (version update from 10.2 to 11.4 - again a different major release) as the following error shows
message: [CompVSharedLib] Failed to load library with path=/home/nvidia/axess_eval/ultimateALPR-SDK-master/binaries/jetson/aarch64/libultimatePluginTensorRT.so, Error: libcudart.so.11.0: cannot open shared object file: No such file or directory

Finally, some additional words on JetPack 4.6.2:
JetPack 4.6.2 is the latest and final release for the TX2 family. In addition, Nvidia made a significant step by moving from TensorRT 7 to TensorRT 8 regardi

Below the full output

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

sudo chmod +x ./prepare.sh && sudo ./prepare.sh

Your Jetpack version is Version: 4.6-b199
Version: 4.6-b197
Preparing for Jetpack 4.4.1...
*[ULTALPR_SDK INFO]:
** This application is used to generate optimized NVIDIA TensorRT models **

*[COMPV INFO]: [UltAlprSdkEngine]Call: optimizeTRT
*[COMPV INFO]: [CompVBase] Initializing [base] modules (v 1.0.0, nt 1)...
*[COMPV INFO]: [CompVBase] sizeof(compv_scalar_t)= #8
*[COMPV INFO]: [CompVBase] sizeof(float)= #4
*[COMPV INFO]: Initializing window registery
*[COMPV INFO]: [ImageDecoder] Initializing image decoder...
*[COMPV INFO]: [CompVCpu] H: '', S: '', M: '', MN: 'ARMv8 Processor rev 3 (v8l)'
*[COMPV INFO]: [CompVBase] CPU features: [arm];[arm64];neon;neon_fma;vfpv4;
*[COMPV INFO]: [CompVBase] CPU cores: online=#6, conf=#6
*[COMPV INFO]: [CompVBase] CPU cache1: line size: #64B, size :#0KB
*[COMPV INFO]: [CompVBase] CPU Phys RAM size: #3831GB
*[COMPV INFO]: [CompVBase] CPU endianness: LITTLE
*[COMPV INFO]: [CompVBase] Binary type: AArch64
*[COMPV INFO]: [CompVBase] Intrinsic enabled
*[COMPV INFO]: [CompVBase] Assembler enabled
*[COMPV INFO]: [CompVBase] Code built with option /arch:NEON
*[COMPV INFO]: [CompVBase] OS name: Jetson_TFTRT
*[COMPV INFO]: [CompVBase] Math Fast Trig.: true
*[COMPV INFO]: [CompVBase] Math Fixed Point: true
*[COMPV INFO]: [CompVMathExp] Init
*[COMPV INFO]: [CompVBase] Default alignment: #32
*[COMPV INFO]: [CompVBase] Best alignment: #32
*[COMPV INFO]: [CompVBase] Heap limit: #262144KB (#256MB)
*[COMPV INFO]: [CompVParallel] Initializing [parallel] module...
*[COMPV INFO]: [CompVParallel] [Parallel] module initialized
*[COMPV INFO]: [CompVBase] [Base] modules initialized
*[COMPV INFO]: [UltAlprSdkTRT] Optimizing models in [../../../assets/models.tensorrt/] folder...
*[COMPV INFO]: [FileUtils] Loading files in ../../../assets/models.tensorrt/ ...
*[COMPV INFO]: [CompVSharedLib] Loading sharded library from /home/nvidia/axess_eval/ultimateALPR-SDK-master/binaries/jetson/aarch64/libultimatePluginTensorRT.so
***[COMPV ERROR]: function: "open()"
file: "/home/ultimate/compv/base/compv_sharedlib.cxx"
line: "71"
message: [CompVSharedLib] Failed to load library with path=/home/nvidia/axess_eval/ultimateALPR-SDK-master/binaries/jetson/aarch64/libultimatePluginTensorRT.so, Error: libnvinfer.so.7: cannot open shared object file: No such file or directory
***[COMPV ERROR]: function: "open()"
file: "/home/ultimate/compv/base/compv_sharedlib.cxx"
line: "73"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "newObj()"
file: "/home/ultimate/compv/base/compv_sharedlib.cxx"
line: "119"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "optimize()"
file: "/home/ultimate/ultimateALPR/SDK_dev/lib/source/ultimate_alpr_sdk_trt.cxx"
line: "78"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "optimizeTRT()"
file: "/home/ultimate/ultimateALPR/SDK_dev/lib/source/ultimate_alpr_sdk_private_engine.cxx"
line: "733"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "UltimateBaseDebugPrint()"
file: "/home/ultimate/ultimateALPR/SDK_dev/lib/../../../ultimateBase/lib/include/ultimate_base_debug.h"
line: "51"
message: [[Check Code]]Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) -> OptimizeTRT failed
***[ULTALPR_SDK ERROR]: function: "main()"
file: "/home/ultimate/ultimateALPR/SDK_dist/samples/c++/trt_optimizer/trt_optimizer.cxx"
line: "61"
message:

*** Failed: code -> 20003, phrase -> OptimizeTRT failed ***

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Martin

[Mamadou DIOP]

Hello,

Ok, that's version incompatibility issue. You'll need to upgrade or downgrade the OS on the device 

Sent from my Galaxy

-------- Original message --------

From: Martin Lechner <martin....@tuwien.ac.at>

Date: 4/18/24 10:12 (GMT+01:00)

To: doubango-ai <douba...@googlegroups.com>

Subject: Re: Difference between "free" binaries available on Github and purchasing a "license"

Hello,

let me step in here for my colleague to give you details about our issue.
Yes, we have tried it on JetPack 4.6.2, and the main issue is that Nvidia went from TensorRT 7 (7.1.3) in 4.4.1 to TensorRT 8 (8.2.1) in 4.6.2. As you already said, the SKD will run on other versions of Jetpack when at least the major TensorRT and CUDA match. Just for reference, we also tried running the 5.1.0 implementation without success.

Some more details:
On Jetpack 4.6.2, the compilation will fail with the message

message: [CompVSharedLib] Failed to load library with path=/home/nvidia/axess_eval/ultimateALPR-SDK-master/binaries/jetson/aarch64/libultimatePluginTensorRT.so, Error: libnvinfer.so.7: cannot open shared object "Courier New">message: [CompVSharedLib] Failed to load library with path=/home/nvidia/axess_eval/ultimateALPR-SDK-master/binaries/jetson/aarch64/libultimatePluginTensorRT.so, Error: libcudart.so.11.0: cannot open shared object "open()"
"/home/ultimate/compv/base/compv_sharedlib.cxx"
line: "71"
message: [CompVSharedLib] Failed to load library with path=/home/nvidia/axess_eval/ultimateALPR-SDK-master/binaries/jetson/aarch64/libultimatePluginTensorRT.so, Error: libnvinfer.so.7: cannot open shared object "open()"

"/home/ultimate/compv/base/compv_sharedlib.cxx"
line: "73"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "newObj()"

"/home/ultimate/compv/base/compv_sharedlib.cxx"
line: "119"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "optimize()"

"/home/ultimate/ultimateALPR/SDK_dev/lib/source/ultimate_alpr_sdk_trt.cxx"
line: "78"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "optimizeTRT()"

"/home/ultimate/ultimateALPR/SDK_dev/lib/source/ultimate_alpr_sdk_private_engine.cxx"
line: "733"
message: Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) ->
***[COMPV ERROR]: function: "UltimateBaseDebugPrint()"

"/home/ultimate/ultimateALPR/SDK_dev/lib/../../../ultimateBase/lib/include/ultimate_base_debug.h"
line: "51"
message: [[Check Code]]Operation Failed (COMPV_ERROR_CODE_E_NOT_FOUND) -> OptimizeTRT failed
***[ULTALPR_SDK ERROR]: function: "main()"

To view this discussion on the web visit https://groups.google.com/d/msgid/doubango-ai/f1c9b20d-3627-4dac-85f0-6616f1225bd4n%40googlegroups.com.