Dynamic Subspace

Category: Computers

  • Finding the Best Bang for the Buck in Generative AI Hardware

    Desktop PC with NVIDIA RTX 3090 Founders Edition GPU

    As I documented last year, I made a substantial investment in my computer workstation for doing local text and image generative AI work by upgrading to 128GB DDR4 RAM and swapping out a RTX 3070 8GB video card for NVIDIA’s flagship workstation card, the RTX A6000 48GB video card.

    After I used that setup to help me with editing the 66,000 word Yet Another Science Fiction Textbook (YASFT) OER, I decided to sell the A6000 to recoup that money (I sold it for more than I originally paid for it!) and purchase a more modest RTX 4060 Ti 16GB video card. It was challenging for me to justify the cost of the A6000 when I could still work, albeit more slowly, with lesser hardware.

    Then, I saw Microcenter begin selling refurbished RTX 3090 24GB Founder Edition video cards. While these cards are three years old and used, they sell for 1/5 the price of an A6000 and have nearly identical specifications to the A6000 except for having only half the VRAM. I thought it would be slightly better than plodding along with the 4060 Ti, so I decided to list that card on eBay and apply the money from its sale to the price of a 3090.

    As you can see above, the 3090 is a massive video card–occupying three slots as opposed to only two slots by the 3070, A6000, and 4060 Ti shown below.

    RTX 4060 Ti
    RTX 3070 8GB
    AMD Ryzen 7 System with NVIDIA A6000 Video Card

    The next hardware investment that I plan to make is meant to increase the bandwidth of my system memory. The thing about generative AI–particularly text generative AI–is the need for lots of memory and more memory bandwidth. I currently have dual-channel DDR4-3200 memory (51.2 GB/s bandwidth). If I upgrade to a dual-channel DDR5 system, the bandwidth will increase to a theoretical maximum of 102.4 GB/s. Another option is to go with a server/workstation with a Xeon or Threadripper Pro that supports 8-channel DDR4 memory, which would yield a bandwidth of 204.8 GB/s. Each doubling of bandwidth roughly translates to doubling how many tokens (the constituent word/letter/punctuation components that generative AI systems piece together to create sentences, paragraphs, etc.) are output by a text generative AI using CPU + GPU inference (e.g., llama.cpp). If I keep watching for sales, I can piece together a DDR5 system with new hardware, but if I want to go with an eight-channel memory system, I will have to purchase the hardware used on eBay. I’m able to get work done so I will keep weighing my options and keep an eye out for a good deal.

  • “With cheap processors . . . what can’t we do?”

    Altair 8800 running at the Southeast Vintage Computer Festival in Atlanta, Georgia in 2014.
    Altair 8800 kit computer running at the SEVCF 2014. This computer is mentioned in Buchanan’s article.

    In the November 1977 issue of Analog Science Fiction/Science Fact magazine, Martin Buchanan published a feature article on personal computers titled, “Home Computers Now!” In it, he opens with a scenario about how PCs can automate family life and then goes into the nuts and bolts of how computers work, what to look for in a kit, and what the future of computing looks like. It was at the end of the article that this passage stood out to me:

    "With cheap processors, cheap memory, and cheap communications, what can't we do? The effects on individuals and society will be major and unpredictable. Today's personal computer is just a beginning" (Buchanan 74).


    Buchanan, Martin. “Home Computers Now!” Analog, Nov. 1977, pp. 61-74.

  • How I Guide Stable Diffusion with ControlNet and Composite Images

    GIMP showing a multi-layer image of Lynn Conway on the right and her co-authored textbook Introduction to VLSI Systems on the left.

    For the illustration of Lynn Conway and her co-authored textbook Introduction to VLSI Systems at the top of yesterday’s post, I used a locally hosted installation of Automatic1111’s stable-diffusion-webui, the finetuned model Dreamshaper 5, which is based on StabilityAI’s Stable Diffusion 1.5 general model, and the ControlNet extension for A1111.

    Stable Diffusion is an image generating AI model that can be utilized with different software. I used Automatic1111’s stable-diffusion-webui to instruct and configure the model to create images. In its most basic operation, I type into the positive prompt box what I want to see in the output image, I type into the negative prompt box what I don’t want to see in the output image, and click “Generate.” Based on the prompts and default parameters, I will see an image output on the right that may or may not align with what I had in mind.

    Automatic1111's stable-diffusion-webui image generating area

    For the positive prompt, I wrote: 

    illustration of a 40yo woman smiling slightly with a nervous expression and showing her teeth with strawberry-blonde hair and bangs, highly detailed, next to a textbook titled introduction to VLSI systems with microprocessor circuits on the cover, neutral background, <lora:age_slider_v6:1>

    I began by focusing on the type of image (an illustration), then describing its subject (woman), other details (the textbook), and the background (neutral). The last part in angle brackets is a LoRA or low rank adaptation. It further tweaks the model that I’m using, which in this case is Dreamshaper 5. This particular LoRA is an age slider, which works by inputting a number that corresponds with the physical appearance of the subject. A “1” presents about middle age. A higher number is older and a lower/negative number is younger.

    Automatic1111's stable-diffusion-webui ControlNet extension area

    ControlNet, which employs different models focused on depth, shape, body poses, etc. to shape the output image’s composition, is an extension to Automatic1111’s stable-diffusion-webui that helps guide the generative AI model to produce an output image more closely aligned with what the user had in mind.

    For the Lynn Conway illustration, I used three different ControlNet units: depth (detecting what is closer and what is further away in an image), canny (one kind of edge detection for fine details), and lineart (another kind of edge detection for broader strokes). Giving each of these different levels of importance (control weight) and telling stable-diffusion-webui when to begin using a ControlNet (starting control step) and when to stop using a ControlNet (ending control step) during each image creation changes how the final image will look.

    Typically, each ControlNet unit uses an image as input for its guidance on the generative AI model. I used the GNU Image Manipulation Program (GIMP) to create a composite image with a photo of Lynn Conway on the right and a photo of her co-authored textbook on the left (see the screenshot at the top of this post). Thankfully, Charles Rogers added his photo of Conway to Wikipedia under a CC BY-SA 2.5 license, which gives others the right to remix the photo with credit to the original author, which I’ve done. Because the photo of Conway cropped her right arm, I rebuilt it using the clone tool in GIMP.

    I input the image that I made into the three ControlNets and through trial-and-error with each unit’s settings, A1111’s stable-diffusion-webui output an image that I was happy with and used on the post yesterday. I used a similar workflow to create the Jef Raskin illustration for this post, too.

  • Lynn Conway, the VLSI Revolution, and Hacking Pedagogy

    Illustration of Lynn Conway and a copy of her textbook with Carver Mead: Introduction to VLSI Systems. Image created with Stable Diffusion.
    Illustration of Lynn Conway with a copy of Mead and Conway’s Introduction to VLSI Systems. Conway’s likeness is based on Charles Roger’s photo on Wikipedia, which he released under a CC BY-SA 2.5 License. Image created with Stable Diffusion.

    This past weekend, The New York Times ran an obituary for Lynn Conway, half of the namesake for the Mead-Conway VLSI Revolution and co-author of the groundbreaking textbook Introduction to VLSI Systems (1980). She died at the age of 86.

    What is so cool about the Mead-Conway VLSI chip design revolution was not only that it was the paradigm shift that made possible the next step in microprocessor design and fabrication by enabling electrical engineering and computer science students to do the work that was previously the domain of physicists and the high tech industry, but also that it was a under-the-radar pedagogical hack. Conway writes in the October 2018 issue of Computer:

    "With all the pieces in place, an announcement was made on ARPANET to electrical engineering and computer science departments at major research universities about what became known as "MPC79." On the surface, while appearing to be official and institutionally based, it was done in the spirit of a classic "MIT hack"--a covert but visible technical stunt that stuns the pubic, who can't figure out how it was done or what did it. (I'd been an undergrad at MIT in the 1950s).

The bait was the promise of chip fabrication for all student projects. Faculty members at 12 research universities signed on to offer Mead-Conway VLSI design courses. This was bootleg, unofficial, and off the books, underscoring the principle that "it's easier to beg forgiveness than to get permission" (p. 69).

    While this was a huge contribution to the development of the computer industry leading into the 1980s and beyond, it was only one of her many accomplishments–innovating an out-of-order queuing processing system for IBM only to be fired in 1968 when she began transitioning to become a woman, starting her career over and eventually making her way to Xerox PARC, later joining the University of Michigan as a professor of electrical engineering and computer science and serving as associate dean of engineering, and becoming a transgender advocate later in life. She was recognized with many awards and honorary doctorates for her contributions to the field as an engineer and educator.

  • Syntrend Creative Park in Taipei, Taiwan

    Syntrend Creative Park in Taipei, Taiwan

    Y sent me this photo of Syntrend Creative Park in Taipei, Taiwan. The whole complex is the funky, weird-ass building in the foreground and the highrise in the background.

    It’s kind of like a more integrated version of Brooklyn’s Industry City + Microcenter area.

    Syntrend has across its many floors shopping (computers, gaming, pop culture, toys, music, etc.) and dining experiences. It has maker spaces, co-working spaces, and event spaces. It’s definitely on my list of places to visit when I get a chance to go to Taiwan again!