[HN Gopher] Future Chips Will Be Hotter Than Ever
       ___________________________________________________________________
        
       Future Chips Will Be Hotter Than Ever
        
       Author : voxadam
       Score  : 20 points
       Date   : 2025-04-16 14:36 UTC (2 hours ago)
        
 (HTM) web link (spectrum.ieee.org)
 (TXT) w3m dump (spectrum.ieee.org)
        
       | FirmwareBurner wrote:
       | Pentium 4, GeForce FX 5800, PS3, Xbox 360, Nintendo Wii, MacBook
       | 20??-2019: _" First time?"_
        
         | TMWNN wrote:
         | _mfw you forget AMD Thunderbird_
         | 
         | Sometimes the solution is worse than the problem. My favorite
         | example is the TRS-80 Model II and its descendants, with the
         | combination of the fan and disk drives so loud that users
         | experience physical discomfort.
         | <https://archive.org/details/80-microcomputing-
         | magazine-1983-...>
        
           | FirmwareBurner wrote:
           | Modern computers should come with built in haptic and rumble
           | motors that can emulate HDD, FDD and CD-ROM sounds whenever
           | you start a game or app. Change my mind.
        
             | TMWNN wrote:
             | <https://tryklack.com/>
             | 
             | But this only simulates keyboard and mouse click sounds. In
             | any case, you wrote "whenever you _start_ a game or app "
             | (my emphasis). The Model II's fan _and drive_ noises are
             | 100% present from start to finish, with the combination
             | enough to drive users insane (or, at least, not want to use
             | the $5-10,000 computer).
        
         | the__alchemist wrote:
         | This checks out. If y'all haven't specced a modern PC: Coolers
         | for GPU and CPU are huge, watercooling is now officially
         | recommended for new CPUs, and cases are ventilated on all
         | sides. Disk bays are moved out of the main chamber to improve
         | airflow. Fans everywhere. Front panels surface areas are
         | completely covered in fans.
        
           | tempodox wrote:
           | Will there be an official "cleared for frying eggs" badge?
           | We'll have to do _something_ with all that heat.
        
           | arcanemachiner wrote:
           | > watercooling is now officially recommended for new CPUs
           | 
           | First I'm hearing of this. Last I checked, air coolers had
           | basically reached parity with any lower-end water cooled
           | setup.
        
             | the__alchemist wrote:
             | I was surprised too, but that's from the AMD label!
        
             | giantg2 wrote:
             | I built a PC last year and saw a bunch of the CPUs were
             | recommending water cooling. There were a few high end air
             | coolers that were compatible. I went with an AIO water
             | cooler. It was cheap and easy. It should give as good or
             | better temperature control as the air coolers that are 5x
             | more expensive.
             | 
             | My guess is manufacturers don't want to tell people they
             | should air cool if it requires listing specific models.
             | It's easy to just say they recommend water cooling since
             | basically all water coolers will provide adequate
             | performance.
        
         | rayiner wrote:
         | The most power hungry P4 didn't top 115W.
        
           | FirmwareBurner wrote:
           | Which was huge in the era when coolers looked like this:
           | https://www.newegg.com/cooler-master-air-cooler-
           | series-a73/p...
        
       | trehalose wrote:
       | > In a frontside design, the silicon substrate can be as thick as
       | 750 micrometers. Because silicon conducts heat well, this
       | relatively bulky layer helps control hot spots by spreading heat
       | from the transistors laterally. Adding backside technologies,
       | however, requires thinning the substrate to about 1 mm to provide
       | access to the transistors from the back.
       | 
       | This is a typo here, right? 1mm is thicker, not thinner, than 750
       | micrometers. I assume 1um was meant?
        
       | nottorp wrote:
       | <looks at the arm macs> You sure?
        
         | Almondsetat wrote:
         | Are you dismissing a technical article with detailed
         | explanations and arguments about the future of CPUs by simply
         | mentioning some piece of current consumer hardware?
        
           | nottorp wrote:
           | Yes, because I think they're extatic about going the wrong
           | way.
        
         | pier25 wrote:
         | The Apple Silicon chips are indeed running hotter on every new
         | generation, no?
        
       | Workaccount2 wrote:
       | Whose gonna pull the trigger on beryllium oxide mounting packages
       | first?
       | 
       | Its the holy grail of having thermal conductivity somewhere
       | between aluminum and copper, while being as electrically
       | insulating as ceramic. You can put the silicon die directly on
       | it.
       | 
       | Problem is that the dust from it is terrifyingly toxic, but in
       | it's finished form it's "safe to handle".
        
         | giantg2 wrote:
         | Won't you have conductivity issues if the oxide layer is
         | damaged?
        
         | pitaj wrote:
         | > Problem is that the dust from it is terrifyingly toxic, but
         | in it's finished form it's "safe to handle".
         | 
         | Doesn't that mean it would be problematic for electronics
         | recycling?
        
       | onewheeltom wrote:
       | Seems my M1 Macbook Air generates almost no heat.
        
         | m463 wrote:
         | good for a laptop. what would the clocks be on a desktop part
         | that was liquid cooled?
        
       ___________________________________________________________________
       (page generated 2025-04-16 17:00 UTC)