We have added a CPU-only processing mode. For users that do not have a compatible Graphics Card (GPU). This option is only available in A.I. Gigapixel 1.1 or newer, can can be found adjacent to the GPU processing option. CPU processing has the following features:
Please note: If your machine hardware falls between the Minimum and Recommended performance levels, you WILL encounter issues using A.I. Gigapixel, including excessive processing times! Reduce the size of your image, or install hardware that meets or exceeds the performance of Recommended hardware, if you encounter problems.
Unsupported Graphics Cards
Please note: If you are using a graphics card that is on the Unsupported Graphics Cards list, then your graphics card hardware will not be supported. This means A.I. Gigapixel may or may not work on your system when using GPU processing within A.I. Gigapixel. We recommend using CPU-only processing if your card is on this list:
Click to view list
Extreme Graphics 1/2
GMA Series (G43/G45/etc)
HD Graphics - Ironlake (Clarkdale)
HD Graphics - Ironlake (Arrandale)
HD Graphics - Sandy Bridge
HD Graphics - Ivy Bridge
HD Graphics 2000
HD Graphics 2500
HD Graphics 3000
HD Graphics P3000
HD Graphics 4000
HD Graphics 4400
HD Graphics 4600
HD Graphics P4600
HD Graphics P4700
GeForce 6000 series
GeForce 7000 series
GeForce 8000 series
GeForce 9000 series
GeForce 100 series
GeForce 200 series
GeForce 300 series
GeForce 400 series
GeForce 500 series
GeForce 600 series
GeForce 700m series
GeForce 800m series
All NVIDIA cards with <2GB of dedicated VRAM are unsupported
Radeon X1000 series
Radeon HD 2000 series
Radeon HD 3000 series
Radeon HD 4000 series
Radeon HD 5000 series
Radeon HD 6000 series
Radeon HD 7000 series
Mobility Radeon X1000 series
Mobility Radeon HD 2000 series
Mobility Radeon HD 3000 series
Mobility Radeon HD 4000 series
Mobility Radeon HD 5000 series
Mobility Radeon HD 6000 series
Mobility Radeon HD 7000 series
Mobility Radeon HD 8000 series
All AMD cards with <2GB of dedicated VRAM are unsupported
Operating System Requirements
|Operating System Version||
|Display Size||800 x 600||1024 x 640|
|OpenGL Support Level What's this?||OpenGL 3.3|
OpenGL (Open Graphics Library) is the computer industry's standard application program interface ( API ) for defining 2-D and 3-D graphic images. Prior to OpenGL, any company developing a graphical application typically had to rewrite the graphics part of it for each operating system platform and had to be cognizant of the graphics hardware as well. With OpenGL, an application can create the same effects in any operating system using any OpenGL-adhering graphics adapter.
OpenGL specifies a set of "commands" or immediately executed functions. Each command directs a drawing action or causes special effects. A list of these commands can be created for repetitive effects. OpenGL is independent of the windowing characteristics of each operating system, but provides special "glue" routines for each operating system that enable OpenGL to work in that system's windowing environment. OpenGL comes with a large number of built-in capabilities requestable through the API. These include hidden surface removal, alpha blending (transparency), antialiasing , texture mapping, pixel operations, viewing and modeling transformations, and atmospheric effects (fog, smoke, and haze).
Graphics Card Requirements
Click here to identify your Graphics Card
Open About Window
Select the Apple icon on the far left side of your menu bar, and select About this Mac
|NVIDIA||GeForce GTX 770 2GB||GeForce GTX 960 4GB||GeForce GTX 1070 Ti 4GB|
|AMD||Radeon HD 8570 2GB||Radeon R9 270 4GB||RX 470 8GB or better|
|Intel||HD Graphics 5000||Iris Plus Graphics 640||N/A|
Due to the power demands of A.I. Gigapixel, we STRONGLY encourage users to process on a desktop with an adequate graphics card.
|NVIDIA||GeForce GTX 860M 2GB||GeForce GTX 1070M 4GB||GeForce GTX 1080M 8GB|
|AMD||R7 M360 2GB||R9 M460 4GB||RX/Radeon Pro 560 8GB|
|Intel||Iris Plus Graphics 655||N/A||N/A|
I'm still confused!
If you've read the requirements here, and don't see your card, don't know how your card compares, or if it's even supported, there are a few things you can do.
Note: We will not be answering "is my card good?" questions, due to the sheer number of hardware combinations that are possible, and our limited support capacity. Thanks!
1. Compare your card
Let's say you follow the advice above to identify your card, but you don't see it mentioned anywhere on this page. What do you do? Compare it to a graphics card we do list.
Comparing graphics cards is easy! There are a few different websites that do these comparisons for you. When you get to one of these comparison pages, you'll see a LOT of numbers, percentages, and stats. These are meant to give some depth to the performance profile of a graphics card, rather than simply saying, "yeh, it's fast" or "that's a good graphics card". Most of the information won't be very useful, but there is one measure that does help -- overal performance. This will tell you, generally, how well a card compares to another. Here is a sample comparison between our Recommended NVIDIA desktop graphics card (NVIDIA GTX 960 4GB) and our Optimal NVIDIA Graphics Card (NVIDIA GTX 1070 Ti 8GB):
RESULT: The NVIDIA GTX 1070 Ti 8GB has an effective speed of 183% (2.8x) more than the GTX 960 4GB. This is an overall comparison based on the benchmarks provided by UserBenchmark.com
As promised, here are a list of sites that do these comparisons for you:
2. My card isn't on the list and it isn't clear if it's supported!
Unfortunately, the determination of support for a card is going to be performance based. There is no hard and set list of graphics cards that will definitively work and a list of graphics cards that will not work. At this time, we do not actively block access to any graphics cards, so theoretically, if your graphics card meets the OpenGL 3.3 requirement and you are using a supported OS version, you CAN INSTALL. You will quickly discover, with A.I. Gigapixel, if you hardware is up to the task.
3. I can install, but I can't figure out if my card is supported, and A.I. Gigapixel crashes!
In this case, you may need to wait for a compatibility update we have planned for A.I. Gigapixel that will expand the supported machine criteria by allowing processing to take place in your CPU, only. This will help those of you that have Intel Graphics Cards, only, use the product. This implementation will be tremendously slower than the GPU implementation, so if you're seeing crashes and have a strong CPU, our advice is to compare what you're currently using with a card with performance at a level that we recommend, and invest in hardware to run this tool. There will likely be no reduction in demand that this software puts on your hardware, so your hardware will need to be improved to a level that allows A.I. Gigapixel to operate without issue.
Can't you just tell me what to buy?
We do not recommend specific graphics cards to buy. Sorry!
Approximate Calculation Values
A.I. Gigapixel is an extremely powerful tool, primarily because it performs TRILLIONS of calculations on your image while upsampling. In order to enable the state-of-the-art results that A.I. Giapixel provides, your system will be asked to make so many calculations, your head could spin! Here's a simple guide to estimate how many calculations are made for each pixel in your image:
1.4 million per pixel
4x & 6x
2.4 million per pixel
What does this mean?
Well, put simply, the larger your image and the larger the upsampling, the more calculations your machine will have to do. This is unavoidable, and will result in cases where processing can take quite a long time. Here are some rough estimates of the number of calculations that will take place when you resize various input image sizes at different scales.
|Image Size vs Scaling||Pixel Count||2x
100% - 200%
200% - 400%
400% - 600%
|1000 x 1000||1 million||1.2 trillion||2.4 trillion||2.4 trillion|
|1920 x 1080 (HQ)||2.1 million||2.4 trillion||4.9 trillion||4.9 trillion|
|3840 x 2160 (4k)||8.2 million||9.9 trillion||19.9 trillion||19.9 trillion|
|5120 x 2880 (5k)||14.7 million||17.6 trillion||35.3 trillion||35.3 trillion|
|7680 x 4320 (8k)||33.1 million||39.8 trillion||79.6 trillion||79.6 trillion|