Last Updated on
RetroArch can run on the usual platforms like Windows, Mac OS X and Linux, but it stands alone in that it can support far more platforms beyond just that.
We support operating systems that not even Microsoft and Apple themselves support anymore, such as macOS X on PowerPC Macs, and RetroArch being available on Windows OSes as far back as Windows 95.
On top of all that, RetroArch also runs on iOS and Android for tablets and phones, as well as on game consoles like PS2, PS3, PSP, PS Vita, Wii, Wii U, 2DS, 3DS, Switch, and more!
RetroArch is available on many popular store fronts for free, including Steam, Amazon App Store, Huawei App Gallery, and Google Play!
The current stable version is: 1.9.10
High Dynamic Range fixes for D3D11/D3D12!
- Fixed contrast to be more correct – now scales from 0-10 linearly and behaves more the way you’d expect it to – changed name to ditch legacy settings users may have
- D3D11/HDR: Fixed D3D11’s blend, rasterizer and topology states not being set to the sames when using HDR and leaving the menu – caused issues with PCSX2’s Shadow of the Colossus
- Added ability to skip inverse tonemapper to the shader via the constant buffer using ‘inverse_tonemap’ – set to 0.0f to skip
- Fixed potential bug when swapping between HDR and SDR and the bit depth not being set correctly
Filters
1.9.10 adds a new Picoscale_256x-320×240 video filter. This uses a number of high quality, high performance algorithms developed by irixxxx for Picodrive standalone to upscale 256×224, 256×239 and 256×240 content to 320×240 (content of any other resolution is passed through unchanged).
Much like the existing Upscale_256x-320×240 filter, this is intended for use on platforms/devices with native 320×240 resolution support, where it greatly reduces aliasing while producing a significantly sharper image than conventional (hardware) bilinear filtering.
Three filter variants are provided:
Picoscale_snn_256x-320×240: ‘Smoothed’ nearest neighbour
Picoscale_bl2_256x-320×240: 2-level-bilinear with 2 quantized weights
Picoscale_bl4_256x-320×240: 4-level-bilinear with 4 quantized weights
Essentially, both image ‘smoothness’ and performance requirements increase in order of snn -> bl2 -> bl4.
Go here to see several screenshots demonstrating the output of each filter type: (click for full-size images)
These filters are highly efficient. Tested with the Snes9x 2005 Plus core (a lightweight core which is nonetheless at the upper limit of many low powered handheld devices), we see the following increase in total performance overheads when each filter is applied (note that the existing Upscale_256x-320×240 filter is included for comparison):