TurboCAD Forums

The Ultimate Resource for TurboCAD Knowledge

Register
 
When replying to a specific statement be sure to quote the previous post.

Graphics Card Recommendations?
Read 11432 times
October 08, 2014, 11:51:13 PM
When I look at technical sites such as Tom's Hardware for graphics cards recommendations, they admit on the first page that they orient their selections for the benefit of gamers.  I don't play FPS games, but Turbocad does use 3D graphics, so it seems probable that the better cards for video games would be better cards for Turbocad.

Is the above assumption correct?

That being the case, which features benefit Turbocad?  On-board memory (if so, how much)? A specific number of channels? DirectX, Shader or OpenGL support? Multiple GPUs?  Clock speed or memory speed?

On the other hand, which features are not used by Turbocad or are just not worth the money?

Thanks in advance,

Jeff

Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


* October 12, 2014, 10:27:05 AM
#1
Hi Jeff
Depending on which version your using I recall when installing Tc there txt. file that mention recommendations for graphic cards that is suitable.The file some where in the program file I think.Its been a while.Your comment in the third paragraph IMO the features plays a part in Tc. for better rendition.You raise no easy answer it all depends on how Tc is used.I would think exploring graphic cards that support CAD software would be option rather gamers but I had success with both.

W.D.

Logged
Even a mistake may turn out to be the one thing necessary to a worthwhile achievement Henry Ford
If I have ever made any valuable discoveries it has been owing more to patient attention than to any other talent Isaac Newton
I have not failed Ive just found 10,000 ways that won't work Thomas Edison


* October 12, 2014, 08:55:38 PM
#2

That being the case, which features benefit Turbocad?  On-board memory (if so, how much)? A specific number of channels? DirectX, Shader or OpenGL support? Multiple GPUs?  Clock speed or memory speed?

On the other hand, which features are not used by Turbocad or are just not worth the money?

Thanks in advance,

Jeff

http://www.redway3d.com/pages/GPUList.php

This link may be of some help to you

Logged
DonCW
2017 Pro with Light Works Plug in
2018 Platinum

Windows 10
There's so much to learn and not much time left to learn it.


October 14, 2014, 11:39:16 PM
#3

http://www.redway3d.com/pages/GPUList.php

This link may be of some help to you

Thanks, but I've reviewed that list and it did not specify which cards were better than others, let alone explain why.  Also, I am interested in the possible performance improvements for non-RedSDK rendering modes.


Regards,

Jeff

Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


October 16, 2014, 08:46:47 PM
#4
Jeff:
This is a difficult subject. You can tell by the number of reads, and lack of responses. I have attempted to get helpful information about this subject for years, from this forum and many other places. The problem(s) associated with giving advice or recommendations about the technical performance for the variety of functions a modern video card is required to perform regarding a specific program, are legion.
Hardware and software and driver interactions are amazingly complex; add to that all the ways individuals can apply these resources to any given task, and you end up like the centipede laying in the ditch trying to figure out how to walk. That said, I will attempt to elucidate my experience and {limited!!} understanding of this subject.
Your first question first: “That being the case, which features benefit Turbocad?”
Rule of thumb, from old school hardware; Ya never can have too much RAM. Anytime your system has to perform a “disk cache” operation, it will slow down performance SIGNIFICANTLY. If you have sufficient on-board ram and wide data pathways on your hardware (motherboard), you will experience much better performance overall. It presents itself even more when you are working with large files and large data set transfers. So the answer to your second question: “(if so, how much)?” is dependent on the size of the files and programs you will be running, and the size of the data sets you will be manipulating while doing so. Hardware speed, i.e. CPU speed, data buss speed, RAM speed, etc. will all play a huge role in the overall performance of your system. Getting all the parts and pieces working together at optimal speeds will always provide the best performance of any system you have. (I did say it was complex!  )
You ask about “A specific number of channels?” I am assuming you are talking about dual or quad channel Mother Board channels. Here again, the more the merrier. A quad channel system will run (IF it is configured with the correct population of RAM, and the proper CPU) will always beat the crap out of any dual channel system. So if performance is your goal, then get a quad channel system and populate it with as much RAM as you can afford, and the fastest multi-core CPU as you can afford.
ALWAYS always always get a system that supports Direct X and Open GL, and as many current technologies as possible.  Keep your options as wide as possible.
Multiple GPU’s may, or may not be a good solution for you, depending on what you are going to use your system for. The key is to select all the components for your system to work together to take advantage of each individual component’s strengths.  My strategy is to invest in a platform that allows me the best performance for the dollar I spend NOW, and to leave options for future development of components that most likely will improve the system I now have. A tough thing to accomplish, since my crystal ball has a big crack in it. Clock speed and memory speed are again, one of those things you want to marry so the total system works together for the best overall performance. Getting one or the other, or even both, without matching the components so they ALL work together without ‘bottlenecks’ will do little good. It is the SYSTEM that provides performance.
Which brings me to your last question:  “On the other hand, which features are not used by Turbocad or are just not worth the money?”
What features of TC are YOU going to use? That is the real question. If you want all possible applications of TC to work as fast as possible, then a healthy dose of unobtainium should be on your shopping list.
My understanding of how TC works and what I want to do with it (as well as several other programs I use extensively) is what I used to put together my shopping list for the computer I now use. As of today, I am still not sure I made all the right choices. I probably could have saved some money and gotten very near the same performance; but there ARE features on the hardware I selected that serve my purpose for TC and the other applications I use. I am sure that there are more learned folks out there that will correct my suppositions, and I welcome any enlightenment that has factual support.
TC uses CPU and RAM for rendering.
TC uses GPU for manipulating wire frame, hidden line, and draft renders. And yes, it uses CPU and RAM as well for these functions, but the BOTTLENECK is usually the GPU for these functions.
If you do not have enough RAM, TC will use disk cache to store data while it renders, slowing everything down to a crawl. If you have narrow data pathways (dual vs quad channel) then it will also slow things down. If you have a slow CPU, and/or few cores, then it will also take more time to render. TC does take advantage of multi-core/multi-thread architecture for rendering.
For the GPU and Video card: There are two distinct applications for graphic cards:   1)Games.  2)Everything else.  At least, that is what the literature and blogs and opinionaters would lead us to believe. The truth is somewhere in between. The vid card I now have, kills on the games I play (I am not an avid gamer, however!). I am sure there are better gaming cards, for less money than I spent. But, none of those gaming cards could manipulate my draft renders, hidden line and wire frame drawings anywhere as smooth and fast as the one I have, for anywhere near the same money.
There are a million tons of opinions and discourses out there to wade through, if you are so inclined. I have sampled a few thousand of them over the last few years. I am of the opinion that even a modest amount of money spend on a system that is balanced and well thought out will return excellent performance.  Research, and a sound understanding of what YOU want to be able to do is key to any successful adventure into building your system.
I doubt I answered your questions as well as you would like. I hope I helped some, anyway!

Bud

Logged
Win 8.1 Pro ~TC 21.2 Pro Plat
ASUS Sabertooth X79 LGA 2011 Intel X79
Intel Core i7-4930K~PNY NVIDIA Quadro K2200
G.SKILL Ripjaws Z Series 32GB (4 x 8GB) DDR3 1866 
Corsair CPU Cooler H100i
SanDisk Ultra Plus 256GB SATA 6.0GB/s SSD
Seagate Barracuda 1 TB (x2)
Stable @ 4.5 GigHz overclocked
NICE


October 17, 2014, 12:36:45 AM
#5
You ask about “A specific number of channels?” I am assuming you are talking about dual or quad channel Mother Board channels.

Nope... talking about the number of 'lines' that the graphic card can use.  The terminology varies between manufacturers; pipelines, CUDA cores, etc.

For the GPU and Video card: There are two distinct applications for graphic cards:   1)Games.  2)Everything else.  At least, that is what the literature and blogs and opinionaters would lead us to believe. The truth is somewhere in between. The vid card I now have, kills on the games I play (I am not an avid gamer, however!). I am sure there are better gaming cards, for less money than I spent. But, none of those gaming cards could manipulate my draft renders, hidden line and wire frame drawings anywhere as smooth and fast as the one I have, for anywhere near the same money.

Bud,

What, specifically, did you look for in the graphic card that made it perform better than the others?


Regards,

Jeff


Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


October 17, 2014, 02:52:59 AM
#6
Jeff,

Specifically, I was looking for a graphics card that has the strength to perform heavy lifting in the area of operations per second, bandwidth, and memory capacity. There are many cards that out-perform the one I ended up choosing, but the price for them was outside my budget. The NVIDIA Quadro K2200 is a cutting edge card that performs spectacularly for the price bracket, has outstanding specifications, and now that I have it in my setup, I have found it shines in all areas I have the greatest demand for performance in. I could do better I am sure, but for the price and my limits at this point in time, I am satisfied.

The main areas I felt would provide the best performance were in the number of CUDA Cores: 640, which is significant. This is where the utility of working in Hidden Line and Draft Render is outstanding.
As valuable as CUDA cores are is GPU Memory: 4GB GDDR5, which is double the GPU RAM of the K2000 this card replaces. The RAM is significant in performance levels and this card maintains a steady hand in dealing with the workloads I throw at it.

Additionally, the Memory Interface is 128-bit, and the Memory Bandwidth is 80 GB/s, which handles a very heavy load with ease. There are times when I overload the card, but so far, it has not crashed or locked up my system. For most drawings and workloads, it is as if I were manipulating cards in solitaire.

The System Interface is PCI Express 2.0 x16; I would like to see this card in the PCIE 3.0 configuration, but that will be a ways out. Perhaps next year.

The Maximum Power Consumption is only 68 W, and it does not need a power jumper from the PSU. It runs on native power from the mother board.

This card has Display Connectors DVI-D DL + DP 1.2 + DP 1.2 which includes the DisplayPort with Audio, which suits me just fine.

The final straw that makes me smile is the 3 Year Warranty. Nothing gives you confidence like a strong warranty.

So, in short, there was nothing specific, but everything in the package that made me choose this card over many other possibilities. As far as I am concerned, this card is a killer deal for the dollar.

Hope this helps,
Bud

Logged
Win 8.1 Pro ~TC 21.2 Pro Plat
ASUS Sabertooth X79 LGA 2011 Intel X79
Intel Core i7-4930K~PNY NVIDIA Quadro K2200
G.SKILL Ripjaws Z Series 32GB (4 x 8GB) DDR3 1866 
Corsair CPU Cooler H100i
SanDisk Ultra Plus 256GB SATA 6.0GB/s SSD
Seagate Barracuda 1 TB (x2)
Stable @ 4.5 GigHz overclocked
NICE


October 17, 2014, 10:15:55 AM
#7
Bud,

Thanks for the criteria.  I wasn't considering a card that expensive because I thought all of those were designed for gamers, but your choice does seem appealing (especially the low power requirements).


Jeff


Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


October 17, 2014, 10:24:36 AM
#8
Did a search for "Comparison Workstation Graphics Cards" and came up with this article.  I am pointing to a page in the article because the first paragraph is very interesting.

http://www.tomshardware.com/reviews/best-workstation-graphics-card,3493-6.html

Got some reading to do this weekend.  Thanks to Bud for reminding me that the term "workstation" is very useful when limiting searches.


Jeff

Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


October 18, 2014, 08:48:13 PM
#9
Found a good (if slightly old) list of tests for graphics cards here: http://www.tomshardware.com/charts/workstation-graphics-2013/benchmarks,146.html

The problem is, different cards are at the top of different test results, with some $150 cards significantly outperforming $1,500 cards in more than a few of them.

My primary applications are mechanical design for production (hidden line and ‘standard’ views) and photorealistic renderings of various objects.

Obviously, some of the tests will not translate well to predicting TurboCAD’s performance, but I’m not so sure which of these tests give the best indication thereof.

So… which tests do you think are best suited for choosing a graphics card for use with TurboCAD?

01 AutoCAD 2013 - 2D - Ortho Lines
02 AutoCAD 2013 - 2D - Radial Plines
03 AutoCAD 2013 - 2D - Text and Blocks
04 AutoCAD 2013 - 2D - Erase / Zoom
05 AutoCAD 2013 - 2D - Performance Summary
06 AutoCAD 2013 - 3D - Rotate Wireframe
07 AutoCAD 2013 - 3D - Rotate Hidden
08 AutoCAD 2013 - 3D - Rotate Conceptual
09 AutoCAD 2013 - 3D - Rotate Realistic
10 AutoCAD 2013 - 3D - Performance Summary
11 Maya 2013 - OpenGL - Werewolf Duett
12 Maya 2013 - OpenGL - Hand
13 Maya 2013 - OpenGL - Insect
14 Maya 2013 - OpenGL - Toy Store
15 OpenGL - SPECViewperf11 - Catia-03
16 OpenGL - SPECViewperf11 - EnSight-04
17 OpenGL - SPECViewperf11 - Lightwave-01
18 OpenGL - SPECViewperf11 - Maya-03
19 OpenGL - SPECViewperf11 - Pro/ENGINEER-05
20 OpenGL - SPECViewperf11 - Solidworks-03
21 OpenGL - SPECViewperf11 - TCVIS-02
22 OpenGL - SPECViewperf11 - SNX-02
23 OpenGL - Synthetik - Unigine Heaven 4.0
24 OpenGL - Synthetik - Unigine Sanctuary
25 OpenGL - Synthetik - Unigine Tropics
26 OpenGL - Synthetik - GPU Caps PostFX
27 OpenGL - Synthetik - Tessmark (OGL 4.0)
28 Autodesk Inventor 2013 - 1000 cubes
29 CUDA - 3ds MAX iRay
30 CUDA - Blender
31 CUDA - Octane DL
32 CUDA - Octane PMC
33 CUDA - Octane PT
34 CUDA - Fluidmark 1080p Score
35 CUDA - Fluidmark 1080p FPS
36 OpenCL - Bitmining
37 OpenCL - Luxmark 2.0 Sala
38 OpenCL - RatGPU Renderer
39 BasemarkCL Image Processing - Blur
40 BasemarkCL Image Processing - Noise Reduction
41 BasemarkCL Image Processing - Surface Smoothing
42 BasemarkCL Image Processing - Sharpening
43 BasemarkCL Video Processing - Blur
44 BasemarkCL Video Processing - Noise Reduction
45 BasemarkCL Video Processing - Surface Smoothing
46 BasemarkCL Video Processing - Sharpening
47 BasemarkCL -Fluid Operation
48 BasemarkCL - Wave Operations
49 BasemarkCL - Juilia Rendering
50 BasemarkCL - Mandelbrot Rendering
51 OpenCL- Folding@Home - FP32- Explicit Solvent
52 OpenCL- Folding@Home - FP32- Implizit Solvent
53 OpenCL- Folding@Home - FP64- Explicit Solvent
54 OpenCL- Folding@Home - FP64- Implizit Solvent
55 Solidworks 2013 OpenGL - Graphics Composite
56 Solidworks 2013 OpenGL - RealView Graphics Composite
57 Solidworks 2013 OpenGL - Shadows Composite
58 Solidworks 2013 OpenGL - Ambient Occlusion Composite
59 Solidworks 2013 OpenGL - Shaded Mode Composite
60 Solidworks 2013 OpenGL - Shaded With Edges Mode Composite
61 Solidworks 2013 OpenGL - Real View Disabled Composite
62 OpenCL Financial Operations - Binominal Option Pricing (FP32)
63 OpenCL Financial Operations - Monte Carlo Option Pricing (FP32)
64 OpenCL Financial Operations - Binominal Option Pricing (FP64)
65 OpenCL Financial Operations - Monte Carlo Option Pricing (FP64)
66 OpenCL AES256 SHA2-256 Bandwidth Enryption/Decryption
67 OpenCL AES256 SHA2-256 Hashing Bandwidth

Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


October 19, 2014, 04:32:27 PM
#10
Did a bit more research and compared a high-end gaming card against a mid-range workstation graphics card.  The comparison is between a GeForce GTX 760 and a Quadro K2000, both with 2 GB of memory.

The Quadro has a clear advantage with OpenGL applications (11 of the 18 tests where both cards were tested), but falls well short for CUDA and OpenCL (not the same as OpenGL).  It does okay in AutoCAD 3D (except for 3D Rotate Realistic and Wireframe, where it is smoked by the GTX 760).

All of the cards in the link I previously posted tested pretty much the same for AutoCAD 2D, whether they cost $1,999 or $87.

The results are interesting, but they still don't tell me what I need to know.  I know that RedSDK is designed to work with OpenGL, but RedSDK is an option in TurboCAD.  Which rendering engine does TC use when it is not set to RedSDK?


Jeff


Benchmark Scores (GTX 760 runs away with it)               GeForce GTX 760     Quadro K2000  Units         Ratio
35 CUDA - Fluidmark 1080p FPS                                            49                8 FPS             613%
34 CUDA - Fluidmark 1080p Score                                        3012              500 Score           602%
51 OpenCL- Folding@Home - FP32- Explicit Solvent                      19.34             4.45 ns/day          435%
25 OpenGL - Synthetik - Unigine Tropics                               116.9             31.7 Score           369%
41 BasemarkCL Image Processing - Surface Smoothing                       11                3 Score           367%
36 OpenCL - Bitmining                                                 102.4             28.5 MHashes/s       359%
66 OpenCL AES256 SHA2-256 Bandwidth Enryption/Decryption               5.64             1.57 GB/s            359%
65 OpenCL Financial Operations - Monte Carlo Option Pricing           60.05            16.79 kOpt/s          358%
64 OpenCL Financial Operations - Binominal Option Pricing (           31.08              8.7 kOpt/s          357%
62 OpenCL Financial Operations - Binominal Option Pricing (           82.03             23.3 kOPT/s          352%
45 BasemarkCL Video Processing - Surface Smoothing                       84               24 Score           350%
67 OpenCL AES256 SHA2-256 Hashing Bandwidth                            4.86             1.39 GB/s            350%
37 OpenCL - Luxmark 2.0 Sala                                            502              144 Score           349%
31 CUDA - Octane DL                                                      58              201 Sec             347%
53 OpenCL- Folding@Home - FP64- Explicit Solvent                       3.81              1.1 ns/day          346%
33 CUDA - Octane PT                                                     207              716 Sec             346%
32 CUDA - Octane PMC                                                    235              807 Sec             343%
24 OpenGL - Synthetik - Unigine Sanctuary                             119.3             34.8 FPS             343%
52 OpenCL- Folding@Home - FP32- Implizit Solvent                      76.71            22.59 ns/day          340%
54 OpenCL- Folding@Home - FP64- Implizit Solvent                        4.7             1.39 ns/day          338%
63 OpenCL Financial Operations - Monte Carlo Option Pricing         1623.66           494.82 kOpt/s          328%
27 OpenGL - Synthetik - Tessmark (OGL 4.0)                            21126             6870 Score           308%
26 OpenGL - Synthetik - GPU Caps PostFX                                 132               44 Score           300%
38 OpenCL - RatGPU Renderer                                              52              156 Sec             300%
23 OpenGL - Synthetik - Unigine Heaven 4.0                             40.1             13.5 FPS             297%
09 AutoCAD 2013 - 3D - Rotate Realistic                                1926              693 Score           278%
42 BasemarkCL Image Processing - Sharpening                             255              104 Score           245%
46 BasemarkCL Video Processing - Sharpening                             906              399 Score           227%
50 BasemarkCL - Mandelbrot Rendering                                    962              435 Score           221%
39 BasemarkCL Image Processing - Blur                                   321              146 Score           220%
48 BasemarkCL - Wave Operations                                         855              416 Score           206%
06 AutoCAD 2013 - 3D - Rotate Wireframe                                 739              360 Score           205%
43 BasemarkCL Video Processing - Blur                                  1091              552 Score           198%
49 BasemarkCL - Juilia Rendering                                       1791              913 Score           196%
40 BasemarkCL Image Processing - Noise Reduction                        402              205 Score           196%
16 OpenGL - SPECViewperf11 - EnSight-04                               50.15            27.84 Score           180%
29 CUDA - 3ds MAX iRay                                                  487              816 Sec             168%
44 BasemarkCL Video Processing - Noise Reduction                       1566             1032 Score           152%
28 Autodesk Inventor 2013 - 1000 cubes                                   22               15 Score           147%
47 BasemarkCL -Fluid Operation                                          453              309 Score           147%
10 AutoCAD 2013 - 3D - Performance Summary                             1446             1015 Score           142%
30 CUDA - Blender                                                    196.48           249.07 Sec             127%


Benchmark Scores (Pretty much the same)                    GeForce GTX 760     Quadro K2000  Units         Ratio
08 AutoCAD 2013 - 3D - Rotate Conceptual                               1767             1526 Score           116%
07 AutoCAD 2013 - 3D - Rotate Hidden                                   1541             1482 Score           104%
04 AutoCAD 2013 - 2D - Erase / Zoom                                     484              481 Score           101%
02 AutoCAD 2013 - 2D - Radial Plines                                    643              641 Score           100%
01 AutoCAD 2013 - 2D - Ortho Lines                                      744              744 Score           100%
05 AutoCAD 2013 - 2D - Performance Summary                              438              439 Score           100%
03 AutoCAD 2013 - 2D - Text and Blocks                                  240              242 Score            99%


Benchmark Scores (Quadro K2000 is clear winner)            GeForce GTX 760     Quadro K2000  Units         Ratio
13 Maya 2013 - OpenGL - Insect                                        59.56            43.74 Sec              73%
11 Maya 2013 - OpenGL - Werewolf Duett                                65.31            47.37 Sec              73%
14 Maya 2013 - OpenGL - Toy Store                                    238.34           159.89 Sec              67%
17 OpenGL - SPECViewperf11 - Lightwave-01                             22.02             56.6 Score            39%
15 OpenGL - SPECViewperf11 - Catia-03                                 15.24            42.85 Score            36%
20 OpenGL - SPECViewperf11 - Solidworks-03                            14.98            48.13 Score            31%
18 OpenGL - SPECViewperf11 - Maya-03                                  16.35            53.12 Score            31%
12 Maya 2013 - OpenGL - Hand                                         335.21            48.87 Sec              15%
22 OpenGL - SPECViewperf11 - SNX-02                                    3.02            32.51 Score             9%
19 OpenGL - SPECViewperf11 - Pro/ENGINEER-05                            1.7            19.07 Score             9%
21 OpenGL - SPECViewperf11 - TCVIS-02                                  1.02            36.85 Score             3%


Benchmark Scores (GTX 760 wasn't tested)                   GeForce GTX 760     Quadro K2000  Units
55 Solidworks 2013 OpenGL - Graphics Composite                            0             5.37 Score
56 Solidworks 2013 OpenGL - RealView Graphics Composite                   0             5.64 Score
57 Solidworks 2013 OpenGL - Shadows Composite                             0              5.6 Score
58 Solidworks 2013 OpenGL - Ambient Occlusion Composite                   0             6.67 Score
59 Solidworks 2013 OpenGL - Shaded Mode Composite                         0             5.31 Score
60 Solidworks 2013 OpenGL - Shaded With Edges Mode Composit               0             5.44 Score
61 Solidworks 2013 OpenGL - Real View Disabled Composite                  0             4.44 Score

Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


* October 19, 2014, 10:47:38 PM
#11
Get a card with HDMI.  I have a AMD Radeon HD 7800 Series and it will do DirectX and Open GL along with 1080P, 3D, 4K UHDTV, and 2 GB ram, etc.  I can use it on a 55" LED TV like my other computer that has an AMD Radeon HD 7700 Series.  Picture on a 55" is good.  My cards aren't the higher priced but I shopped for the specs and got within my budget.

Link:  http://www.amd.com/en-gb/products/graphics/desktop/7000/7800#

Good luck and happy shopping,

John B.

Logged
John B.
TCv2019 Platinum
Windows 10 Pro 64 Bit
1 GB Hard Drive 16 GB RAM
AMD Radeon RX 580 Series


October 20, 2014, 09:55:07 AM
#12
John,

Every card I am considering has HDMI.

What concerns me is getting the best performance from TurboCAD.  The links I posted previously show that, for some 3D applications, gaming cards are vastly superior, but for others, workstation cards are the way to go.

Unfortunately, I don't know which of the benchmark tests would most closely apply to TurboCAD (I'm hoping an administrator or other IMSI employee in a position to know will weigh in).


Jeff

Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1


* October 20, 2014, 11:50:12 PM
#13
Very interesting thread.

I recently purchased a Geforce GTX 750 Ti gaming card with 640 cuda cores.  I have not even had the time to plug it in yet, so I can't give you any specific feedback on the card.

What I can say is that I actually supprised myself by the purchase.  I thought I was going to buy a 770 or 780 until I heard that the next version of the gaming card due out fairly soon will be much more powerful than the current versions, and that the 750Ti was actually a very efficient card.
Not so much of a powerhouse though, but the number of cuda cores for the price was not that bad.
 
I paid just under $160 for the 750Ti and will wait for the gen 6 processors from intel before I upgrade to a new machine and a high end card.  My card has 2GB of ram, from what I was told if you want to run a 4K monitor then you might want to step up to the 760 GTX with 4GB of ram or a similar CAD card.

I appreciate the info presented here.  I would also like to point out that not all CAD cards are the same.  Some are good at some things and others are good at other things.  I am also currious how all of this applies to the features in TC.  Hope someone will weigh in on this.


Logged
TC V21 Pro Platinum  Win 7 Professional


October 21, 2014, 10:13:12 AM
#14
The NVIDIA Quadro K2200 is a cutting edge card that performs spectacularly for the price bracket, has outstanding specifications, and now that I have it in my setup, I have found it shines in all areas I have the greatest demand for performance in.

Bud,

The RedSDK site doesn't list the Quadro K2200 as a supported card.  Which video driver are you using?  Also, are you using RedSDK or GDI for your renders?  Also (again), would you say most of your work is 2D or 3D?


Thanks in advance,

Jeff

Logged
TC Pro Platinum 2019, 2018, 2017, 2016 & 2015 (all with LightWorks & RedSDK) & V21
System: i7-5820K @ 3.30GHz, ASRock X99 Extreme4, 16GB DDR4-2133 RAM, Gigabyte GTX 970, Samsung NVMe SSD 950 (256GB), Windows 7 Pro (64-bit) SP1