Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register

Quick Links


Shells Virtual Desktop
BMail.ag - Secure Email Service
Server.net
CPLicense.net
VPS Server
Buy VPN
Vultr
VMs for AI
HostDare
ReliableSite White-Label Dedicated Hosting for Resellers
InterServer VPS
BMail.ag - Secure Email Service
Best VPN
High-Performance Bare Metal Server Solutions
Karvl.com
Server Mania Cloud Hosting
DataWagon Hosting
AlphaVPS Hosting
Evoxt.com
Clouvider
Godlike VPS
VPS Hosting with NVMe
Residential IPs in the US & 4G Mobile Proxies in EU & US with Unlimited Bandwidth
ReliableSite White-Label Dedicated Hosting for Resellers
Shells Virtual Desktop

Categories

In this Discussion

Home › Announcements
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

2025 Black Friday / Cyber Monday: FLASH SALE & MEGATHREAD - "The Trade War"

1…692693695697698…733

Comments

  • WolfWolf Member

    @SashkaPro said:

    @jnd said: But if you max out each core individually then you will hit thermal limits or cache/memory sharing issues and the whole cpu will throttle down

    So it is impossible to "predict" what GB6 will be on that VMs?
    I am asking because a see a lot of yabs's with ~500-600 GB6 single core score on 26xx v4 and I just do not understand why ~1100 on empty node and ~500 on fully (?) booked node... It is okay if on full node gb6 will be ~900, okay ~850 but not x2.

    Many factors come into play here.

    Multiple active cores will cause a slower clock speed (also Cores vs. Threads) and the Caches (L3) will be split between all instances. (~35 MB / 56 Threads) Also there is Memory Bandwith, Thermals and more...

    Thanked by 5SashkaPro tux tentor jnd FAT32
  • AlteredParadoxAlteredParadox Member

    Calling it good enough for now, lol

    Thanked by 1Wolf
  • AlteredParadoxAlteredParadox Member

    @Cybr said:

    @AlteredParadox said:
    killing powersave helps a tad, but not amazing. Looks like dartnode sets the fan mode to optimal and doesn't give operator/admin to ipmi to change it.

    Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1181                          
Multi Core      | 9019                          
Full Test       | https://browser.geekbench.com/v6/cpu/15401210

    If they thermally throttle my 9950X I'm gonna be mad as hell.

    I wrote off iKoula after the high end dedi I got from them was thermally throttled due to poor cooling, and they just ignored my ticket completely.

    ipmitool still totally works to change the fan speeds, so less of a problem - digging deeper it definitely wasn't thermal throttling anyway tho.

  • JordJord Moderator, Host Rep, Megathread Squad

    5 more pages :D

    Thanked by 1FAT32
  • AlteredParadoxAlteredParadox Member

    JORD

    Thanked by 1Jord
  • AlteredParadoxAlteredParadox Member

    @Jord

    Thanked by 1Jord
  • AlteredParadoxAlteredParadox Member

    WELCOME

  • JordJord Moderator, Host Rep, Megathread Squad

    Morning, thank you @AlteredParadox

  • cosmossofacosmossofa Member

    A little bit late now, but i just wonder, no poll this year ?

    (Nevermind, just a tiny contribution to 700)

  • bestobesto Member

    So it is impossible to "predict" what GB6 will be on that VMs?
    I am asking because a see a lot of yabs's with ~500-600 GB6 single core score on 26xx v4 and I just do not understand why ~1100 on empty node and ~500 on fully (?) booked node... It is okay if on full node gb6 will be ~900, okay ~850 but not x2.

    In VPS, when they say "core" it means "half a core". Basically they are saying one thread. Though it has become standard terminology now. So, you are getting half a cores performance in a single core score.

    Regarding prediction, generally better/latest processors give better score, but if it is VPS, if the server is over-provisioned, then it would have lesser score.

    For most tasks, single thread performance matters much much more than multi-thread because most software cant make use of multi. Only things like video-rendering, or hostings lot of high user websites make good use of multi.

    Thanked by 2jnd boba
  • maxexxmaxexx Member

    Is it still available?

    @AlteredParadox said:
    dartnode 99/yr dedi yabs (upgraded ram and nvme ssd)

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2025-04-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Fri Dec  5 00:46:29 CST 2025

Basic System Information:
---------------------------------
Uptime     : 0 days, 0 hours, 3 minutes
Processor  : Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz
CPU cores  : 56 @ 1200.000 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 125.8 GiB
Swap       : 49.2 GiB
Disk       : 889.4 GiB
Distro     : Debian GNU/Linux 13 (trixie)
Kernel     : 6.12.57+deb13-amd64
VM Type    : NONE
IPv4/IPv6  : ✔ Online / ❌ Offline

IPv4 Network Information:
---------------------------------
ISP        : Snaju Development
ASN        : AS399646 Snaju Development
Host       : Snaju Development
Location   : Houston, Texas (TX)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/nvme1n1p2):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 220.53 MB/s  (55.1k) | 466.67 MB/s   (7.2k)
Write      | 221.11 MB/s  (55.2k) | 469.13 MB/s   (7.3k)
Total      | 441.64 MB/s (110.4k) | 935.81 MB/s  (14.6k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 640.40 MB/s   (1.2k) | 719.96 MB/s    (703)
Write      | 674.42 MB/s   (1.3k) | 767.90 MB/s    (749)
Total      | 1.31 GB/s     (2.5k) | 1.48 GB/s     (1.4k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | 862 Mbits/sec   | 618 Mbits/sec   | --             
Eranium         | Amsterdam, NL (100G)      | 868 Mbits/sec   | 726 Mbits/sec   | --             
Uztelecom       | Tashkent, UZ (10G)        | 761 Mbits/sec   | 427 Mbits/sec   | --             
Leaseweb        | Singapore, SG (10G)       | busy            | busy            | --             
Clouvider       | Los Angeles, CA, US (10G) | 917 Mbits/sec   | 914 Mbits/sec   | --             
Leaseweb        | NYC, NY, US (10G)         | 920 Mbits/sec   | 856 Mbits/sec   | --             
Edgoo           | Sao Paulo, BR (1G)        | 833 Mbits/sec   | 591 Mbits/sec   | --             

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1184                          
Multi Core      | 8224                          
Full Test       | https://browser.geekbench.com/v6/cpu/15399487

YABS completed in 11 min 4 sec
  • _MS__MS_ Member

    Deals?

    Thanked by 1FAT32
  • WolfWolf Member

    @maxexx said:
    Is it still available?

    @AlteredParadox said:
    dartnode 99/yr dedi yabs (upgraded ram and nvme ssd)

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2025-04-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Fri Dec  5 00:46:29 CST 2025

Basic System Information:
---------------------------------
Uptime     : 0 days, 0 hours, 3 minutes
Processor  : Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz
CPU cores  : 56 @ 1200.000 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 125.8 GiB
Swap       : 49.2 GiB
Disk       : 889.4 GiB
Distro     : Debian GNU/Linux 13 (trixie)
Kernel     : 6.12.57+deb13-amd64
VM Type    : NONE
IPv4/IPv6  : ✔ Online / ❌ Offline

IPv4 Network Information:
---------------------------------
ISP        : Snaju Development
ASN        : AS399646 Snaju Development
Host       : Snaju Development
Location   : Houston, Texas (TX)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/nvme1n1p2):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 220.53 MB/s  (55.1k) | 466.67 MB/s   (7.2k)
Write      | 221.11 MB/s  (55.2k) | 469.13 MB/s   (7.3k)
Total      | 441.64 MB/s (110.4k) | 935.81 MB/s  (14.6k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 640.40 MB/s   (1.2k) | 719.96 MB/s    (703)
Write      | 674.42 MB/s   (1.3k) | 767.90 MB/s    (749)
Total      | 1.31 GB/s     (2.5k) | 1.48 GB/s     (1.4k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | 862 Mbits/sec   | 618 Mbits/sec   | --             
Eranium         | Amsterdam, NL (100G)      | 868 Mbits/sec   | 726 Mbits/sec   | --             
Uztelecom       | Tashkent, UZ (10G)        | 761 Mbits/sec   | 427 Mbits/sec   | --             
Leaseweb        | Singapore, SG (10G)       | busy            | busy            | --             
Clouvider       | Los Angeles, CA, US (10G) | 917 Mbits/sec   | 914 Mbits/sec   | --             
Leaseweb        | NYC, NY, US (10G)         | 920 Mbits/sec   | 856 Mbits/sec   | --             
Edgoo           | Sao Paulo, BR (1G)        | 833 Mbits/sec   | 591 Mbits/sec   | --             

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1184                          
Multi Core      | 8224                          
Full Test       | https://browser.geekbench.com/v6/cpu/15399487

YABS completed in 11 min 4 sec

    Nope. Flash-sale in limited quantity.

    Thanked by 2FAT32 maxexx
  • FairShareFairShare Member

    I want Liberty from this LET thread !

    I want this to be closed ASAP ! @FAT32

    Thanked by 1FAT32
  • AlteredParadoxAlteredParadox Member

    nope. we have to make it to 700

    Thanked by 1FAT32
  • AlteredParadoxAlteredParadox Member

    at least 700

  • COLBYLICIOUSCOLBYLICIOUS Member

    I'm in!

  • allthemtingsallthemtings Member, Megathread Squad

    @AlteredParadox said:
    killing powersave helps a tad, but not amazing. Looks like dartnode sets the fan mode to optimal and doesn't give operator/admin to ipmi to change it.

    Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1181                          
Multi Core      | 9019                          
Full Test       | https://browser.geekbench.com/v6/cpu/15401210

    GB6 doesnt scale with that many cores try gb4/5 you see the real power

    Thanked by 1FAT32
  • CybrCybr Member

    @AlteredParadox said:

    @Cybr said:

    @AlteredParadox said:
    killing powersave helps a tad, but not amazing. Looks like dartnode sets the fan mode to optimal and doesn't give operator/admin to ipmi to change it.

    Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1181                          
Multi Core      | 9019                          
Full Test       | https://browser.geekbench.com/v6/cpu/15401210

    If they thermally throttle my 9950X I'm gonna be mad as hell.

    I wrote off iKoula after the high end dedi I got from them was thermally throttled due to poor cooling, and they just ignored my ticket completely.

    ipmitool still totally works to change the fan speeds, so less of a problem - digging deeper it definitely wasn't thermal throttling anyway tho.

    Some CPUs are notorious for throttling down their frequency by insane amounts with a number of high cores, but the E5-2680v4 doesn't seem to be one of them, since according to wikichip it boosts to 2.9ghz with 5-14 cores: https://en.wikichip.org/wiki/intel/xeon_e5/e5-2680_v4

    Thanked by 1FAT32
  • lewellynlewellyn Member

    @allthemtings said:

    @AlteredParadox said:
    killing powersave helps a tad, but not amazing. Looks like dartnode sets the fan mode to optimal and doesn't give operator/admin to ipmi to change it.

    Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1181                          
Multi Core      | 9019                          
Full Test       | https://browser.geekbench.com/v6/cpu/15401210

    GB6 doesnt scale with that many cores try gb4/5 you see the real power

    yabs.sh -9 indeed

    Thanked by 2allthemtings FAT32
  • AlteredParadoxAlteredParadox Member

    now I shall install proxmox, then an opnsense vm, then setup my secondary ip on said opnsense vm, then make multiple interfaces, then make vms, then figure out something to put on the vms!!!! All the things.

  • AlteredParadoxAlteredParadox Member

    Also onward to 700!!!

  • AlteredParadoxAlteredParadox Member

    More deals!

  • AlteredParadoxAlteredParadox Member

    Also it’s 5am and I haven’t slept yet, damnit

  • lewellynlewellyn Member

    @AlteredParadox said:
    Also it’s 5am and I haven’t slept yet, damnit

    Don't feel too bad. It's quarter past 6am here and I have to be up to get ready for work in 15 mins. And I obviously failed the sleep test too...

  • FAT32FAT32 Administrator, Deal Compiler Extraordinaire

    Man it has already been days but I am still feeling very tired... it takes so long to recover from not sleeping straight for 3 days :(

    Thanked by 14truemagic Soli admax Wolf barbarza nullnothere doobydoe cainyxues geo besto boba hzhoanglee JohnnySac vadiheg
  • ralfralf Member

    @FAT32 said:
    Man it has already been days but I am still feeling very tired... it takes so long to recover from not sleeping straight for 3 days :(

    If you're saying you're not able to sleep, the RN thread is in need of more shitposting...

    Thanked by 1FAT32
  • lewellynlewellyn Member
    edited December 2025

    Or wrangle more deals here. There's still fools, er, users who wish to be parted with their money.

  • FAT32FAT32 Administrator, Deal Compiler Extraordinaire

    @ralf said:

    @FAT32 said:
    Man it has already been days but I am still feeling very tired... it takes so long to recover from not sleeping straight for 3 days :(

    If you're saying you're not able to sleep, the RN thread is in need of more shitposting...

    My sleep scheduled is messed up - I can't feel refreshed even after 8 hours of sleep.

    Thanked by 5admax doobydoe tentor geo JohnnySac
  • admaxadmax Member, Megathread Squad

    @FAT32 said:

    @ralf said:

    @FAT32 said:
    Man it has already been days but I am still feeling very tired... it takes so long to recover from not sleeping straight for 3 days :(

    If you're saying you're not able to sleep, the RN thread is in need of more shitposting...

    My sleep scheduled is messed up - I can't feel refreshed even after 8 hours of sleep.

    I've been having nightmares these past few days.

    Thanked by 2FAT32 Murv
This discussion has been closed.