Message boards :
Number crunching :
Help finding a single slot GPU card
Message board moderation
Author | Message |
---|---|
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
Looks like there was another one from ELSA Also a Frankenstein project to modify the reference GTX 750 Ti with a AMD cooler Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
Found this MSI on eBay MSI GeForce GTX 750 Ti DirectX 11.2 N750TI-2GD5/OC It has a double-wide PCIe bracket but from the pictures, a hacksaw or cutting wheel can lop off the unnecessary vent part to single width and not interfere with any components or video output connectors. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Brent Norman Send message Joined: 1 Dec 99 Posts: 2786 Credit: 685,657,289 RAC: 835 |
Obviously not very popular :-(Or went up in a puff of smoke, LOL |
Cruncher-American Send message Joined: 25 Mar 02 Posts: 1513 Credit: 370,893,186 RAC: 340 |
Just check it out for being a 750ti - I bought one on eBay that had a (IIRC) GTS 450 graphics chip in it(!). I found out because I noticed it was running slow, and when I finally took it apart to look (because of the odd characteristics it was showing in PrecisionX) the chip on the board was a GTS 450... It was a pure fake, as the BIOS had been rigged to respond as a 750ti, but the chip was not a 750ti. Buyer beware, indeed. |
Terror Australis Send message Joined: 14 Feb 04 Posts: 1817 Credit: 262,693,308 RAC: 44 |
Does it have to be a GTX-750 ? There's evidently AMD RX-460 and Nvidia GTX-1050 single slot cards available T.A. |
MarkJ Send message Joined: 17 Feb 08 Posts: 1139 Credit: 80,854,192 RAC: 5 |
There was a GTS450SP card. I bought 10 of them a few years ago when I was in Japan for work. They are rather dated now. (Released in 2010). They had 192 CUDA cores and were based on the GF106 chip. They also tend to overheat as the fan can’t move enough air unless you crank it up to 100% due to being a single slot design. BOINC blog |
spitfire_mk_2 Send message Joined: 14 Apr 00 Posts: 563 Credit: 27,306,885 RAC: 0 |
GTX 1030 single slot, 80 USD: https://www.evga.com/products/product.aspx?pn=02G-P4-6338-KR Refurbished (meaning used/returned) GT 740 single slot, 45 USD: https://www.evga.com/products/product.aspx?pn=02G-P4-2742-RX |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13731 Credit: 208,696,464 RAC: 304 |
GTX 1030 single slot, 80 USD: https://www.evga.com/products/product.aspx?pn=02G-P4-6338-KR I wouldn't bother with a GT740 for crunching; however it would be interesting to see how the GT 1030 compares to the GTX 750Ti for power use and performance. I suspect the GTX 750Ti would probably beat it in both instances. Grant Darwin NT |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
I have a GTX 750 Ti in my Linux box and it works well with a 384.90 driver provided by SuSE. Also its temperature running SETI@home is 50 C, lower than the temperature of the GTX 1050 Ti in the Windows PC running the same tasks, which reaches 60 C. Tullio |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13731 Credit: 208,696,464 RAC: 304 |
The 750Ti are great little mid range cards for those on a budget, certainly a lot of bang for your buck. You just need the tuning for SOG work. Check out Shaggie's graphs. Even now, the credit they produce for the power they use they're still in the top section with all of the latest & greatest. Grant Darwin NT |
Brent Norman Send message Joined: 1 Dec 99 Posts: 2786 Credit: 685,657,289 RAC: 835 |
100 more Watts from adding a 750Ti? Nahh, more like 30 at most. |
rob smith Send message Joined: 7 Mar 03 Posts: 22189 Credit: 416,307,556 RAC: 380 |
The addition may have just upset the PSU - some are so badly designed that move away from a sweet-spot and the power-in to power-out ratio goes completely ape (and that change can be an increase or decrease in load power demand) Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13731 Credit: 208,696,464 RAC: 304 |
The power meter says that the second card takes an extra 100w of total PC power, bit low I would have thought, oh well. We'll see how we go. Actually way too high. The GTX 750Ti is rated at 60W (an overclocked after market unit might pull another 10W). But even running SoG with the most aggressive settings, the most mine would pull was around 45W- peak. Grant Darwin NT |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Actually if I study my power meter closely the extra varies from 80-100W.I would begin to wonder if you haven't been sold one of those fakes with an older-generation core chip but a spoofed bios claiming to be a Maxwell. That power draw is above the 75W limit for a PCI slot: I think all 750Tis are slot powered, although some have an power cable for the extra fans. If it is a fake, it'll be less productive in spite of the extra power. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13731 Credit: 208,696,464 RAC: 304 |
Some guys have both separate cards shown, I just have 2 x 980?? It's just the way it gets reported to the servers. If you check out the result file for a WU, it shows the actual model of all cards on that system. Grant Darwin NT |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
My GTX 750 Ti has an extra 6 pin power connector. My GTX 1050 Ti has no extra power connector, but the PSU gives only 300 W, and I suspect this is the cause of frequent reboots of Windows 10 while running GPU tasks. No reboots on the Linux box with its 400 W power supply. I just cleaned its inside with my vacuum cleaner. Tullio |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
OK, that's not looking too bad. I have some known genuine 750 Tis, purchased new when they first came out - the basic features look OK, and the extensions compare like this: RH known genuine CS under investigation ---------------- ---------------------- cl_khr_global_int32_base_atomics cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_byte_addressable_store cl_khr_icd cl_khr_icd cl_khr_gl_sharing cl_khr_gl_sharing cl_nv_compiler_options cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_pragma_unroll cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_d3d11_sharing cl_nv_copy_opts cl_nv_copy_opts cl_nv_create_buffer |
spitfire_mk_2 Send message Joined: 14 Apr 00 Posts: 563 Credit: 27,306,885 RAC: 0 |
My GTX 750 Ti has an extra 6 pin power connector. My GTX 1050 Ti has no extra power connector, but the PSU gives only 300 W, and I suspect this is the cause of frequent reboots of Windows 10 while running GPU tasks. No reboots on the Linux box with its 400 W power supply. I just cleaned its inside with my vacuum cleaner. You could downcloack the card, that way it would need less juice. I have 750 Ti SC, one fan, no additional power connector. Works fine at factory clock. |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
You could downcloack the card, that way it would need less juice. Reboots happen when running SETI@home GPU tasks, not Einstein@home GPU tasks. This maybe because on my AMD A10-6700 which should have 4 cores (but Windows 10 Task Manager says it has two cores and 4 logical processors) I can run two SETI CPU tasks and one GPU task, while it runs one Einstein CPU task and one GPU task because Einstein GPU tasks uses more CPU. Tullio |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.