[johnnym]

I have bought a few additional things one additional machine (the AS 800) during my absence here and my old listing is gone with Nekochan, so I'll create a new list of DEC/Digital hardware I have collected over the years in this thread (I hope I get all data correctly):

• AXPpci 33 (with LCA4 @166 MHz in a nice old generic AT desktop case)
  
• AlphaStation 200 4/166 (with EV4 @166 MHz)
  
• AlphaStation 255/300 (with EV45 @300 MHz)
  
• AlphaStation 500/400 (with EV56 @400 MHz and purchased from FlasBurn (found his username through the Nekonomicon) still on Nekochan)
  
• PWS 500au (with EV56 @500 MHz)
  
• AlphaServer 800 (with EV56 @400 MHz)
  
• AlphaServer DS20e (with 2 x EV67 @666 MHz)
  
• AlphaServer DS25 (with 2 x EV68CB @1 GHz)
  
• 2 x AlphaServer ES45 (each with 4 x EV68CB @1 GHz)
  

All in working order, except for the AlphaStation 500 which only works sometimes, because it has a flaky PSU, which I already knew but bought it anyhow to secure it for the collection.

I recently used one ES45 and the DS25 to produce some packages for OpenBSD 7.1 (which is missing packages for the Alpha architecture). The gcc 8 took an aggregated 16 hours or so on the ES45 and I'm still unsure if it produces correct code. Because I get a weird error when trying to compile devel/ninja (dependency of htop):

PHP Code:

ds25# make clean
===>  Cleaning for ninja-1.10.2p1
ds25# time make -j3 install
===>  Checking files for ninja-1.10.2p1
[...]
c++ -MMD -MT build/deps_log.o -MF build/deps_log.o.d -g -Wall -Wextra -Wno-deprecated -Wno-missing-field-initializers -Wno-unused-parameter -fno-rtti -fno-exceptions -fvisibility=hidden -pipe '-DNINJA_PYTHON="/usr/local/bin/python3.9"' -O2 -DNDEBUG -fdiagnostics-color -I/usr/local/include -DUSE_PPOLL -DNINJA_HAVE_BROWSE -I. -O2 -pipe -c ./src/deps_log.cc -o build/deps_log.o
./src/deps_log.cc: In member function 'LoadStatus DepsLog::Load(const string&, State*, std::__cxx11::string*)':
./src/deps_log.cc:288:1: error: unrecognizable insn:
 }
 ^
(insn 476 475 477 35 (set (reg:DI 263)
        (plus:DI (reg/f:DI 65 virtual-stack-vars)
            (const_int -524292 [0xfffffffffff7fffc]))) "./src/deps_log.cc":210 -1
     (nil))
during RTL pass: vregs
./src/deps_log.cc:288:1: internal compiler error: in extract_insn, at recog.c:2309
libbacktrace could not find executable to open
Please submit a full bug report,
with preprocessed source if appropriate.
See <https://gcc.gnu.org/bugs/> for instructions.
when running:  c++ -MMD -MT build/deps_log.o -MF build/deps_log.o.d -g -Wall -Wextra -Wno-deprecated -Wno-missing-field-initializers -Wno-unused-parameter -fno-rtti -fno-exceptions -fvisibility=hidden -pipe '-DNINJA_PYTHON="/usr/local/bin/python3.9"' -O2 -DNDEBUG -fdiagnostics-color -I/usr/local/include -DUSE_PPOLL -DNINJA_HAVE_BROWSE -I. -O2 -pipe -c ./src/deps_log.cc -o build/deps_log.o
Traceback (most recent call last):
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 519, in <module>
    objs += cxx(name, variables=cxxvariables)
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 287, in cxx
    return n.build(built(name + objext), 'cxx', src(name + '.cc'), **kwargs)
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 169, in build
    self._run_command(self._expand(cmd, local_vars))
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 194, in _run_command
    subprocess.check_call(cmdline, shell=True)
  File "/usr/local/lib/python3.9/subprocess.py", line 373, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command 'c++ -MMD -MT build/deps_log.o -MF build/deps_log.o.d -g -Wall -Wextra -Wno-deprecated -Wno-missing-field-initializers -Wno-unused-parameter -fno-rtti -fno-exceptions -fvisibility=hidden -pipe '-DNINJA_PYTHON="/usr/local/bin/python3.9"' -O2 -DNDEBUG -fdiagnostics-color -I/usr/local/include -DUSE_PPOLL -DNINJA_HAVE_BROWSE -I. -O2 -pipe -c ./src/deps_log.cc -o build/deps_log.o' returned non-zero exit status 1. 


I should maybe try again w/o -j[..] first, because I am reluctant to invest another 16 hours into recompiling gcc 8.

[Irinikus]

I have bought a few additional things during my absence here and my old listing is gone with Nekochan, so I'll create a new list of DEC/Digital hardware I have collected over the years in this thread (I hope I get all data correctly):

• AXPpci 33 (with LCA4 @166 MHz in a nice old generic AT desktop case)
  
• AlphaStation 200 4/166 (with EV4 @166 MHz)
  
• AlphaStation 255/300 (with EV45 @300 MHz)
  
• AlphaStation 500/400 (with EV56 @400 MHz and purchased from a community member still on Nekochan)
  
• PWS 500au (with EV56 @500 MHz)
  
• AlphaServer 800 (with EV56 @400 MHz)
  
• AlphaServer DS20e (with 2 x EV67 @666 MHz)
  
• AlphaServer DS25 (with 2 x EV68CB @1 GHz)
  
• 2 x AlphaServer ES45 (each with 4 x EV68CB @1 GHz)
  

All in working order, except for the AlphaStation 500 which only works sometimes, because it has a flaky PSU, which I already knew but bought it anyhow to secure it for the collection.

I recently used one ES45 and the DS25 to produce some packages for OpenBSD 7.1 (which is missing packages for the Alpha architecture). The gcc 8 took an aggregated 16 hours or so on the ES45 and I'm still unsure if it produces correct code. Because I get a weird error when trying to compile devel/ninja (dependency of htop):

PHP Code:

ds25# make clean
===>  Cleaning for ninja-1.10.2p1
ds25# time make -j3 install
===>  Checking files for ninja-1.10.2p1
[...]
c++ -MMD -MT build/deps_log.o -MF build/deps_log.o.d -g -Wall -Wextra -Wno-deprecated -Wno-missing-field-initializers -Wno-unused-parameter -fno-rtti -fno-exceptions -fvisibility=hidden -pipe '-DNINJA_PYTHON="/usr/local/bin/python3.9"' -O2 -DNDEBUG -fdiagnostics-color -I/usr/local/include -DUSE_PPOLL -DNINJA_HAVE_BROWSE -I. -O2 -pipe -c ./src/deps_log.cc -o build/deps_log.o
./src/deps_log.cc: In member function 'LoadStatus DepsLog::Load(const string&, State*, std::__cxx11::string*)':
./src/deps_log.cc:288:1: error: unrecognizable insn:
 }
 ^
(insn 476 475 477 35 (set (reg:DI 263)
        (plus:DI (reg/f:DI 65 virtual-stack-vars)
            (const_int -524292 [0xfffffffffff7fffc]))) "./src/deps_log.cc":210 -1
     (nil))
during RTL pass: vregs
./src/deps_log.cc:288:1: internal compiler error: in extract_insn, at recog.c:2309
libbacktrace could not find executable to open
Please submit a full bug report,
with preprocessed source if appropriate.
See <https://gcc.gnu.org/bugs/> for instructions.
when running:  c++ -MMD -MT build/deps_log.o -MF build/deps_log.o.d -g -Wall -Wextra -Wno-deprecated -Wno-missing-field-initializers -Wno-unused-parameter -fno-rtti -fno-exceptions -fvisibility=hidden -pipe '-DNINJA_PYTHON="/usr/local/bin/python3.9"' -O2 -DNDEBUG -fdiagnostics-color -I/usr/local/include -DUSE_PPOLL -DNINJA_HAVE_BROWSE -I. -O2 -pipe -c ./src/deps_log.cc -o build/deps_log.o
Traceback (most recent call last):
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 519, in <module>
    objs += cxx(name, variables=cxxvariables)
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 287, in cxx
    return n.build(built(name + objext), 'cxx', src(name + '.cc'), **kwargs)
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 169, in build
    self._run_command(self._expand(cmd, local_vars))
  File "/usr/obj/ports/ninja-1.10.2/ninja-1.10.2/./configure.py", line 194, in _run_command
    subprocess.check_call(cmdline, shell=True)
  File "/usr/local/lib/python3.9/subprocess.py", line 373, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command 'c++ -MMD -MT build/deps_log.o -MF build/deps_log.o.d -g -Wall -Wextra -Wno-deprecated -Wno-missing-field-initializers -Wno-unused-parameter -fno-rtti -fno-exceptions -fvisibility=hidden -pipe '-DNINJA_PYTHON="/usr/local/bin/python3.9"' -O2 -DNDEBUG -fdiagnostics-color -I/usr/local/include -DUSE_PPOLL -DNINJA_HAVE_BROWSE -I. -O2 -pipe -c ./src/deps_log.cc -o build/deps_log.o' returned non-zero exit status 1. 


I should maybe try again w/o -j[..] first, because I am reluctant to invest another 16 hours into recompiling it.

WoW!!! That’s one hell of a collection!

[Linux-RISC]

... a few additional things during my absence here ...

 Excellent collection, congratulations.

[johnnym]

WoW!!! That’s one hell of a collection!

... a few additional things during my absence here ...

 Excellent collection, congratulations.

Thanks, guys. And it took me only about ten years of my life, or so... 

And I have to correct myself: according to my work log, the only machine bought in my absence here, was the AlphaServer 800. My memory tricked me. 

I'd love to have an EV7(z) (e.g. ES47) and a small dual processor one - e.g. DS20L because the DS25 is already a huge machine - but those are too rare now I think and I don't believe I have the adequate funds.