Giovanni's blog

Imagine if mainline linux were a species. Everyone who used it had to upgrade their systems. If they were a procaryote, they'd need to upgrade to a Eukaryote. This is what a version history would look like: 1991 We added a nucleus (MMU). Since 0.01. 1992 64 bit support address space added on DEC Alpha. (Larger genome supported!) 1998 uclinux developed. Loss of function (MMU) + Gain of function (performance/lower latency)   2007 nommu (uclinux) integrated (Mitochondria!) into 2.5.46 2024 RTLinux merged into mainline- an improvement, but RTOSes also offered in nommu systems. 2028 nommu scheduled for removal : (Procaryotes can be procaryotes again.) 203x 32-bit support removal. Sometime. (Divergent evolution). More may be added as the ideas arrive to me.  Halobacteria is a photosynthetic archaea that feeds on salt and light. If photoheterotrophic bacteria can utilize light as an energy source, so can 32-bit and nommu linux machines.  10-3-2026 :  https...

4-23-26: Update I've switched to Bodhi Linux on my Toshiba laptop.                                             

Earlier this week, Linux 7.1 dropped 486 support , as they had previously considered, and while it was long discussed, I can't help but think of this diagram I drew a few months ago  (I updated it slightly for linux 7.1):        32-bit is effectively another species, with little code-compatibility (according to others, apparently). It's the new 8-bit . Since my background was in microbiology, I tend to see code as a lot more mutually intelligible. But I can see why they want to drop 32 bit instructions. They're like sandbags on a hot air balloon. They need a little extra lift. As chipmakers are getting back into memory production, EUV will be an opportunity to develop chips that embed memory in ways that are far more economical. Currently, DRAM is typically integrated on a different chiplet.    As I wrote in the Phoronix forum ,  "My observation is that it would be far more economical for the foundry to integrate both DRAM and CPU on the ...

At a time when mobile devices were becoming even more mainstream, lowering power consumption became a key objective. However, as the Android and iPhone market in 2007 shortly revealed, power consumption took a distant third in priority when it came to having a competitive product. Performance remained the number one improvement, allowing an ever increasing amount of desktop-level apps on a smartphone, as the Symbian era came to a close. A couple years ago, in adding to my research collection of processors that were unique or rare, I stumbled upon the ARM996HS. The ARM996HS: What was it? It is a "clockless" processor in that it is not a synchronous process, but its 100MHz speed may refer to an average speed. Its benefit was offering 1/3rd power consumption of a similar processor, on just 90,000 logic cells. Other processors had been designed to accomplish this, albeit with far more circuitry- the 386SLC , which had around 800,000 transistors (compared to the original 386's...

Today's post (I actually began writing this post on a Wednesday night at 9:51pm) is going to branch out from central processing units and cover two highly interrelated integration steps from the 2010 era, just before the release of the  Intel Graphics Technology . A couple years ago, in March 2024, I did a multi distro test on my Sylvania netbook from 2011, an Intel Atom N450 with 1.66Ghz, 1GB of RAM, NM10 chipset, and an 8GB SATA based SSD card. This era was defined as a transition between the Northbridge based video graphics controllers  (after the AGP based i740 was integrated into the Intel 810 in 1999) and the Intel HD Graphics, which has been a series since 2010 (along with other products such as the Arc). The most common that I found in netbooks and entry-level PCs at the time were the GMA 915, 950 and the GMA 3150 , the latter two which had DirectX 9 support. I had briefly owned the Eee PC 701 4G for a few years shortly after launch, when it cost $200 for the 4GB mo...