@looney2ns, will you please elaborate on why the micro/mini's aren't suitable for running Blue Iris? If I need to dedicate a machine solely to Blue Iris, I'd prefer to tuck it away where it won't be seen and the 1L form factor seems perfect for that.
mat200
IPCT Contributor
- Jan 17, 2017
- 18,921
- 31,639
Typically, those smaller boxes have less powerful CPUs, and issues with installing a 3.5" HDD for video storage and an SSD for the OS.
Remember, many micro boxes also have external power supplies, so take that into consideration in terms of total space needed.
Ah, I didn't realise they wouldn't fit a 3.5" drive. That explains it.
Anyone know the smallest SFF that will fit a 3.5" drive?
looney2ns
IPCT Contributor
FYI on a google AI search asking which of my computers is more efficient running Blue Iris.
I'm only posting here because the link in the search result said to come here for more info
.
Thread has been idle for almost 2 years.
2:35 PM
The Dell Precision 3630 setup is the clear winner for performance, but the OptiPlex 9020 setup will likely draw less power at idle.
Here is the breakdown of how these two systems compare in hardware efficiency, power consumption, and Blue Iris capability.
Hardware & Quick Sync Efficiency
Power Consumption & PSU
Nvidia GPU Utilization
Verdict
12 cameras @ 4MP and h264 with AI duties on 1/2 the cameras
For 12 cameras at 4MP (H.264) with AI duties on 6 of them, the Dell Precision 3630 remains the far superior and more efficient choice overall.
Here is how your specific workload shifts the balance between these two machines.
AI Inference Performance (The Deciding Factor)
Video Streaming & CPU Overhead
Direct Power Comparison for Your Specific Workload
Final Recommendation
Go with the Dell Precision 3630. The minor savings in idle power offered by the old OptiPlex are completely wiped out by the modern architecture, massive AI headroom, and system stability the 3630 provides for a 12-camera setup.
To help optimize this specific deployment, let me know:
I'm only posting here because the link in the search result said to come here for more info
.Thread has been idle for almost 2 years.
2:35 PM
The Dell Precision 3630 setup is the clear winner for performance, but the OptiPlex 9020 setup will likely draw less power at idle.
Here is the breakdown of how these two systems compare in hardware efficiency, power consumption, and Blue Iris capability.
Hardware & Quick Sync Efficiency- Dell 3630 (Winner): Features an 8th/9th Gen Intel CPU. Its Quick Sync video engine handles modern H.265 camera streams flawlessly.
- OptiPlex 9020: Features a 4th Gen Intel CPU. Its older Quick Sync engine lacks efficient native decoding for modern, high-resolution H.265 streams.
Power Consumption & PSU
- OptiPlex 9020 (Lower Idle Draw): The 365W power supply (PSU) and older components will idle around 35W–60W.
- Dell 3630 (Higher Idle Draw): The 800W PSU operates less efficiently at very low loads. Combined with the beefier 3060 Ti, this system will idle higher, around 65W–90W.
- Note: Blue Iris runs 24/7, so idle/low-load power draw is what impacts your electric bill, not the maximum PSU wattage.
Nvidia GPU Utilization
- Nvidia 3060 Ti (Winner): Essential if you plan to use DeepStack or CodeProject.AI for AI person/vehicle detection. It processes alerts in milliseconds.
- EVGA 1060: Capable of basic AI duties but will struggle or slow down if you have many high-megapixel cameras.
- Efficiency Tip: Do not use the Nvidia GPUs for continuous 24/7 video decoding in Blue Iris. It wastes massive amounts of power. Use Intel Quick Sync for decoding and the Nvidia card strictly for AI analysis.
Verdict
- Choose the Dell Precision 3630 if you have more than 6–8 cameras, use H.265 compression, or want fast AI object detection.
- Choose the OptiPlex 9020 only if you want to minimize 24/7 electricity costs and are running a small number of older H.264 cameras without heavy AI needs.
- How many cameras do you have, and what is their resolution?
- Are you planning to use CodeProject.AI for person/vehicle detection?
- Do your cameras shoot in H.264 or H.265 format?
12 cameras @ 4MP and h264 with AI duties on 1/2 the cameras
For 12 cameras at 4MP (H.264) with AI duties on 6 of them, the Dell Precision 3630 remains the far superior and more efficient choice overall.
Here is how your specific workload shifts the balance between these two machines.
AI Inference Performance (The Deciding Factor)
- The Dell 3630 / RTX 3060 Ti Combo: This setup is highly efficient for AI. An RTX 3060 Ti will blaze through your 6 AI-monitored cameras, processing object detection alerts in roughly 10ms to 30ms per image. Because it finishes the work instantly, the card drops right back down to a low-power state.
- The OptiPlex 9020 / GTX 1060 Combo: This setup will bottleneck. A GTX 1060 will take significantly longer—often 80ms to 150ms+ per inference. When multiple cameras trigger simultaneously, the 1060 will get queued up, spike to 100% utilization, stay pinned there longer, and ultimately burn more total energy trying to catch up.
Video Streaming & CPU Overhead
- Total Stream Throughput: Your 12 cameras at 4MP will pump out roughly 30 to 50 Mbps of data depending on frame rates.
- The Dell 3630: Its 8th/9th Gen processor can easily handle this background decoding load via Intel Quick Sync with negligible CPU hit (around 5–10% utilization).
- The OptiPlex 9020: The 4th Gen Haswell CPU will sweat. Even with Quick Sync enabled, a 4th Gen chip processing 48 megapixels of continuous data will see much higher baseline CPU utilization, driving up its "idle" power consumption and leaving less headroom for the rest of the OS.
Direct Power Comparison for Your Specific Workload- At Idle (No Motion): The OptiPlex 9020 might pull 10–15 Watts less at the wall than the 3630 because of its smaller PSU and less complex architecture.
- Under Load (Frequent Motion/AI Triggers): The efficiency flips. The Dell 3630 processes the video and AI alerts so rapidly that its power spikes are brief flashes. The OptiPlex 9020 will experience prolonged power spikes because its CPU and GPU have to work twice as hard for twice as long to clear the same queue.
Final Recommendation
Go with the Dell Precision 3630. The minor savings in idle power offered by the old OptiPlex are completely wiped out by the modern architecture, massive AI headroom, and system stability the 3630 provides for a 12-camera setup.
To help optimize this specific deployment, let me know:
- What AI framework are you using? (e.g., CodeProject.AI, DeepQuest, or built-in Blue Iris tools?)
- Are your cameras using Substreams (low-res streams for continuous viewing and high-res for recording)?
- What frame rate (FPS) are you targeting on the cameras?
