home.social

#decalpha — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #decalpha, aggregated by home.social.

  1. Distracted by DEC Alpha 64-bit microprocessors. #decalpha

  2. Distracted by DEC Alpha 64-bit microprocessors. #decalpha

  3. Distracted by DEC Alpha 64-bit microprocessors. #decalpha

  4. Distracted by DEC Alpha 64-bit microprocessors. #decalpha

  5. Distracted by DEC Alpha 64-bit microprocessors. #decalpha

  6. J'ai une paire de serveurs #decalpha 3000 avec les alimentation HS.
    J'ai les ai testées au préalable déconnectées de l'ordinateur et branchées sur des résistances bobinées (~40W) , une ne démarrait pas et l'autre avait l'air OK. Une fois cette dernière connectée sur l'ordinateur plus de tension ! Après démontage c'est une diode de redressement qui est en court-circuit (et qui ne l'était donc pas lors du test initial)
    J'ai commandé les composants pour réparer ça et j'ai aussi trouvé sur eBay une alim censée fonctionner, les deux devraient traverser l'Atlantique prochainement, à suivre !
    #retrocomputing

  7. J'ai une paire de serveurs #decalpha 3000 avec les alimentation HS.
    J'ai les ai testées au préalable déconnectées de l'ordinateur et branchées sur des résistances bobinées (~40W) , une ne démarrait pas et l'autre avait l'air OK. Une fois cette dernière connectée sur l'ordinateur plus de tension ! Après démontage c'est une diode de redressement qui est en court-circuit (et qui ne l'était donc pas lors du test initial)
    J'ai commandé les composants pour réparer ça et j'ai aussi trouvé sur eBay une alim censée fonctionner, les deux devraient traverser l'Atlantique prochainement, à suivre !
    #retrocomputing

  8. J'ai une paire de serveurs #decalpha 3000 avec les alimentation HS.
    J'ai les ai testées au préalable déconnectées de l'ordinateur et branchées sur des résistances bobinées (~40W) , une ne démarrait pas et l'autre avait l'air OK. Une fois cette dernière connectée sur l'ordinateur plus de tension ! Après démontage c'est une diode de redressement qui est en court-circuit (et qui ne l'était donc pas lors du test initial)
    J'ai commandé les composants pour réparer ça et j'ai aussi trouvé sur eBay une alim censée fonctionner, les deux devraient traverser l'Atlantique prochainement, à suivre !
    #retrocomputing

  9. J'ai une paire de serveurs #decalpha 3000 avec les alimentation HS.
    J'ai les ai testées au préalable déconnectées de l'ordinateur et branchées sur des résistances bobinées (~40W) , une ne démarrait pas et l'autre avait l'air OK. Une fois cette dernière connectée sur l'ordinateur plus de tension ! Après démontage c'est une diode de redressement qui est en court-circuit (et qui ne l'était donc pas lors du test initial)
    J'ai commandé les composants pour réparer ça et j'ai aussi trouvé sur eBay une alim censée fonctionner, les deux devraient traverser l'Atlantique prochainement, à suivre !
    #retrocomputing

  10. J'ai une paire de serveurs #decalpha 3000 avec les alimentation HS.
    J'ai les ai testées au préalable déconnectées de l'ordinateur et branchées sur des résistances bobinées (~40W) , une ne démarrait pas et l'autre avait l'air OK. Une fois cette dernière connectée sur l'ordinateur plus de tension ! Après démontage c'est une diode de redressement qui est en court-circuit (et qui ne l'était donc pas lors du test initial)
    J'ai commandé les composants pour réparer ça et j'ai aussi trouvé sur eBay une alim censée fonctionner, les deux devraient traverser l'Atlantique prochainement, à suivre !
    #retrocomputing

  11. It is as if a dam broke in 2006. Huge architectural diversity before 2004, almost none after 2008.
    #hpc #cpu #x86 #x86_64 #arm #MIPS #SPARC #risc #alpha #decalpha #cray #NEC #itanium #amd64 #SunSPARC

  12. It is as if a dam broke in 2006. Huge architectural diversity before 2004, almost none after 2008.
    #hpc #cpu #x86 #x86_64 #arm #MIPS #SPARC #risc #alpha #decalpha #cray #NEC #itanium #amd64 #SunSPARC

  13. It is as if a dam broke in 2006. Huge architectural diversity before 2004, almost none after 2008.
    #hpc #cpu #x86 #x86_64 #arm #MIPS #SPARC #risc #alpha #decalpha #cray #NEC #itanium #amd64 #SunSPARC

  14. It is as if a dam broke in 2006. Huge architectural diversity before 2004, almost none after 2008.
    #hpc #cpu #x86 #x86_64 #arm #MIPS #SPARC #risc #alpha #decalpha #cray #NEC #itanium #amd64 #SunSPARC

  15. It is as if a dam broke in 2006. Huge architectural diversity before 2004, almost none after 2008.
    #hpc #cpu #x86 #x86_64 #arm #MIPS #SPARC #risc #alpha #decalpha #cray #NEC #itanium #amd64 #SunSPARC

  16. @win8linux I think we had that (whether we wanted it or not) on OpenVMS. At the time I thought it bizarre that a server in the datacentre should run a GUI all the time but given that I ran a GUI at home on an 8 MHz 68k, the AlphaServer could spare the cycles and it was handy for multiple xterm. #vms #decalpha

  17. @win8linux I think we had that (whether we wanted it or not) on OpenVMS. At the time I thought it bizarre that a server in the datacentre should run a GUI all the time but given that I ran a GUI at home on an 8 MHz 68k, the AlphaServer could spare the cycles and it was handy for multiple xterm. #vms #decalpha

  18. @win8linux I think we had that (whether we wanted it or not) on OpenVMS. At the time I thought it bizarre that a server in the datacentre should run a GUI all the time but given that I ran a GUI at home on an 8 MHz 68k, the AlphaServer could spare the cycles and it was handy for multiple xterm. #vms #decalpha

  19. @win8linux I think we had that (whether we wanted it or not) on OpenVMS. At the time I thought it bizarre that a server in the datacentre should run a GUI all the time but given that I ran a GUI at home on an 8 MHz 68k, the AlphaServer could spare the cycles and it was handy for multiple xterm. #vms #decalpha

  20. @win8linux I think we had that (whether we wanted it or not) on OpenVMS. At the time I thought it bizarre that a server in the datacentre should run a GUI all the time but given that I ran a GUI at home on an 8 MHz 68k, the AlphaServer could spare the cycles and it was handy for multiple xterm. #vms #decalpha

  21. A DEC Alpha programmer that hasn't used the #Internet for years getting online and seeing stuff about "alpha programmers" and getting confused because it has nothing to do with the DEC Alpha and it's just a bunch of insecure, misogynistic guys trying to make themselves look better than other people.

    #DECAlpha #DEC #programming #misogyny

  22. A DEC Alpha programmer that hasn't used the #Internet for years getting online and seeing stuff about "alpha programmers" and getting confused because it has nothing to do with the DEC Alpha and it's just a bunch of insecure, misogynistic guys trying to make themselves look better than other people.

    #DECAlpha #DEC #programming #misogyny

  23. A DEC Alpha programmer that hasn't used the #Internet for years getting online and seeing stuff about "alpha programmers" and getting confused because it has nothing to do with the DEC Alpha and it's just a bunch of insecure, misogynistic guys trying to make themselves look better than other people.

    #DECAlpha #DEC #programming #misogyny

  24. A DEC Alpha programmer that hasn't used the #Internet for years getting online and seeing stuff about "alpha programmers" and getting confused because it has nothing to do with the DEC Alpha and it's just a bunch of insecure, misogynistic guys trying to make themselves look better than other people.

    #DECAlpha #DEC #programming #misogyny

  25. CW: on Windows memory allocation

    This blog entry by Raymond Chen, "Why doesn’t Windows use the 64-bit virtual address space below 0x00000000`7ffe0000?"¹ is absolutely brilliant.

    What I thought was a trivial design decision of some kind turns out to have deep roots in 32-bit Windows (the separation between user and kernel) and, this is where it becomes surreal, the DEC Alpha processor :flan_awe:​!

    I quote Raymond Chen directly:

    "On the Alpha AXP, most 32-bit constants can be generated in at most two instructions. But there’s a range of values that requires three instructions: 0x7fff8000 to 0x7fffffff. "

    As the person who ran the first three DEC Alpha systems in the UK outside DEC "REO" (Reading Office) I have a bit of history with these processors and yet I had never ever heard of this.

    Don't ignore the links embedded in the page because they are worth reading too.

    (also, if you do not follow Raymond Chen's blog via RSS you should - his series on processors and the history of Windows design decisions is fabulous).

    #Windows #MemoryAllocation #DECAlpha #WindowsOnAlpha
    __
    ¹ devblogs.microsoft.com/oldnewt

  26. CW: on Windows memory allocation

    This blog entry by Raymond Chen, "Why doesn’t Windows use the 64-bit virtual address space below 0x00000000`7ffe0000?"¹ is absolutely brilliant.

    What I thought was a trivial design decision of some kind turns out to have deep roots in 32-bit Windows (the separation between user and kernel) and, this is where it becomes surreal, the DEC Alpha processor :flan_awe:​!

    I quote Raymond Chen directly:

    "On the Alpha AXP, most 32-bit constants can be generated in at most two instructions. But there’s a range of values that requires three instructions: 0x7fff8000 to 0x7fffffff. "

    As the person who ran the first three DEC Alpha systems in the UK outside DEC "REO" (Reading Office) I have a bit of history with these processors and yet I had never ever heard of this.

    Don't ignore the links embedded in the page because they are worth reading too.

    (also, if you do not follow Raymond Chen's blog via RSS you should - his series on processors and the history of Windows design decisions is fabulous).

    #Windows #MemoryAllocation #DECAlpha #WindowsOnAlpha
    __
    ¹ devblogs.microsoft.com/oldnewt

  27. CW: on Windows memory allocation

    This blog entry by Raymond Chen, "Why doesn’t Windows use the 64-bit virtual address space below 0x00000000`7ffe0000?"¹ is absolutely brilliant.

    What I thought was a trivial design decision of some kind turns out to have deep roots in 32-bit Windows (the separation between user and kernel) and, this is where it becomes surreal, the DEC Alpha processor :flan_awe:​!

    I quote Raymond Chen directly:

    "On the Alpha AXP, most 32-bit constants can be generated in at most two instructions. But there’s a range of values that requires three instructions: 0x7fff8000 to 0x7fffffff. "

    As the person who ran the first three DEC Alpha systems in the UK outside DEC "REO" (Reading Office) I have a bit of history with these processors and yet I had never ever heard of this.

    Don't ignore the links embedded in the page because they are worth reading too.

    (also, if you do not follow Raymond Chen's blog via RSS you should - his series on processors and the history of Windows design decisions is fabulous).

    #Windows #MemoryAllocation #DECAlpha #WindowsOnAlpha
    __
    ¹ devblogs.microsoft.com/oldnewt

  28. CW: on Windows memory allocation

    This blog entry by Raymond Chen, "Why doesn’t Windows use the 64-bit virtual address space below 0x00000000`7ffe0000?"¹ is absolutely brilliant.

    What I thought was a trivial design decision of some kind turns out to have deep roots in 32-bit Windows (the separation between user and kernel) and, this is where it becomes surreal, the DEC Alpha processor :flan_awe:​!

    I quote Raymond Chen directly:

    "On the Alpha AXP, most 32-bit constants can be generated in at most two instructions. But there’s a range of values that requires three instructions: 0x7fff8000 to 0x7fffffff. "

    As the person who ran the first three DEC Alpha systems in the UK outside DEC "REO" (Reading Office) I have a bit of history with these processors and yet I had never ever heard of this.

    Don't ignore the links embedded in the page because they are worth reading too.

    (also, if you do not follow Raymond Chen's blog via RSS you should - his series on processors and the history of Windows design decisions is fabulous).

    #Windows #MemoryAllocation #DECAlpha #WindowsOnAlpha
    __
    ¹ devblogs.microsoft.com/oldnewt

  29. CW: on Windows memory allocation

    This blog entry by Raymond Chen, "Why doesn’t Windows use the 64-bit virtual address space below 0x00000000`7ffe0000?"¹ is absolutely brilliant.

    What I thought was a trivial design decision of some kind turns out to have deep roots in 32-bit Windows (the separation between user and kernel) and, this is where it becomes surreal, the DEC Alpha processor :flan_awe:​!

    I quote Raymond Chen directly:

    "On the Alpha AXP, most 32-bit constants can be generated in at most two instructions. But there’s a range of values that requires three instructions: 0x7fff8000 to 0x7fffffff. "

    As the person who ran the first three DEC Alpha systems in the UK outside DEC "REO" (Reading Office) I have a bit of history with these processors and yet I had never ever heard of this.

    Don't ignore the links embedded in the page because they are worth reading too.

    (also, if you do not follow Raymond Chen's blog via RSS you should - his series on processors and the history of Windows design decisions is fabulous).

    #Windows #MemoryAllocation #DECAlpha #WindowsOnAlpha
    __
    ¹ devblogs.microsoft.com/oldnewt