Test AMD Ryzen 7 1800X, 1700X, 1700 im Test: König in Anwendungen, Prinz in Spielen

@Volker Rißka

Warum benutzt ihr eine alte Version von 3D Particle Movement die bekanntermaßen für Intel CPUs optimiert ist?
Vergleicht eure Ergebnisse mal mit Anandtech:
85866.png
http://images.anandtech.com/graphs/graph11170/85866.png

Solche Diskrepanzen sollten doch mal hinterfragt werden.

Gleiches gilt für Euler3D, bitte entfernt den alten ICC Dispatcher, der AMD CPUs beschneidet. Danke
The Stilt hat dazu was geschrieben:
https://forums.anandtech.com/thread...use-new-thread.2482739/page-224#post-38741728

The only two workloads where removing the dispatcher does make a difference are Caselab Euler3D (ICL from 2009 (?)) and Linpack. In Caselab Euler3D removing the dispatcher improves the performance on AMD CPUs by >30%, while in Linpack the dispatcher doesn't degrade the actual performance but prevents the program from running on AMD CPUs alltogether.
 
T-REX schrieb:
@Volker Rißka

Warum benutzt ihr eine alte Version von 3D Particle Movement die bekanntermaßen für Intel CPUs optimiert ist?
Vergleicht eure Ergebnisse mal mit Anandtech:
Anhang anzeigen 610526
http://images.anandtech.com/graphs/graph11170/85866.png

Solche Diskrepanzen sollten doch mal hinterfragt werden.

Gleiches gilt für Euler3D, bitte entfernt den alten ICC Dispatcher, der AMD CPUs beschneidet. Danke
The Stilt hat dazu was geschrieben:
https://forums.anandtech.com/thread...use-new-thread.2482739/page-224#post-38741728

Bist du Wahnsinnig.

Du kannst hier doch keine Programme posten die Pro AMD sind.. meine güte hast du in denganzen Jahren die du hier bist nichts gelernt :D
 
immortuos schrieb:
Und das ist einfach Schwachsinn, eine GPU wird nur dann zu 100% ausgelastet wenn die CPU genug leistet um sie schnell genug dafür zu füttern.
Wenn er da noch einen alten Q6600 dazugeschmissen hätte wäre die GPU trotz deaktiviertem VSync nicht ausgelastet worden weil der Q6600 einfach zu langsam ist um die Karte zu füttern.

Ok. Das ist schlicht und einfach falsch. Die GPU wird mit Q6600 niemals zu 100% ausgelastet.
 
Da es nicht nur das CPU und das GPU Limit gibt kann man es auch einfacher ausdrücken.
Wenn die GPU Auslastung unter die 95% rutscht besteht eine anderweitige, relevante Limitierung. ;)
 
Deshalb hat Intel Angst vor Zen:

9oVGc83.png

850 points in Cinebench 15 at 30W is quite telling. Or not telling, but absolutely massive. Zeppelin can reach absolutely monstrous and unseen levels of efficiency, as long as it operates within its ideal frequency range.

Zeppelin unterstützt offenbar wie seine Vorgänger cTDP. Setzt man dies auf 30W, zieht das Teil Kreise um alles, was Intel so anbietet; das sind keine Kreise mehr, sondern eher Umlaufbahnen von Planeten ;)

8Rch6JF.png

Leider nimmt die Effizienz bei höherer Taktrate ab. Der 1800X scheint also schon ziemlich ausgereizt.

Warum der 1800X besser ist als der 1700(X) mit OC, und OC generell bei Ryzen eher Nachteile hat:

Due to the seamless integration of the on-die system management unit (SMU, similar in functionality to PMU on Intel) there are quite many differences compared to the previous AMD desktop microarchitectures when it comes to overclocking.

At stock, Ryzen has all of the power management features enabled and the SMU runs the whole operation and is in charge for everything. These power management features ‎enthalten‎ various power, ‎aktuelle‎ and thermal limiters, voltage controllers and power gating features.

All of these can be completely ignored, if you have no plans to overclock the CPU.

For overclocking purposes the engineers at AMD have included a special mode (the "OC Mode"), which will disable all of the limiters, voltage controllers and protections (except the CPU thermal protection) upon the activation.

The "OC Mode" is automatically activated when the user raises the base frequency (P0 PState) of the CPU. The SMU indicates the activation of the "OC Mode" by sending "0C" code to the diagnostic display (Port 80) of the motherboard.

Understanding the different CPU frequency states (PStates), their voltages and especially the actual effective voltage is harder than ever before with Zeppelin. Unlike with the older designs (15h family) the boosted PStates (Turbo & XFR) are completely invisible.

Due to that fact, they are officially called as "Shadow PStates". This means that unlike with the previous designs these PStates are not defined in the standard MSR registers and cannot be modified (or be seen) by the user. The only way the user can even verify their presence is to see them actually firing (i.e. from the actual effective frequency & voltage).

Understanding the voltages specified for the standard PStates can be confusing as well. That's because in the normal operating mode (i.e. "non-OC") the SMU controls the voltages automatically through the voltage controllers.

For example, the P0 PState might specify 1.37500V voltage, while the actual effective voltage during the residency in this state is 1.26250V or slightly higher. This is not a glitch, but the normal operation of the CPU. Basically, the voltage specified in the MSR is just the upper limit and the SMU will automatically add a dynamic negative offset to this value, reducing the actual effective voltage. The amount of the negative offset varies depending on load and the temperature. For the tested sample the offsets were -120mV & -144mV for the two highest base PStates (3.6 & 3.2GHz).

When the "OC Mode" is activated the SMU will disable the voltage controllers, which among other things disables the automatic voltage offsets. This can create an illusion that the power consumption is heavily increasing due to the use of "OC Mode". While technically it is accurate, it is more of a consequence than the actual reason. A vast majority of the increased power consumption comes from the now disabled automatic negative voltage offsets, which causes the actual CPU voltage to increase anything between 50 and 150mV. Because of this behavior, it is advised that the user doesn't increase the CPU voltage right away (when overclocking), but only upon actual demand (as usual).

One of the major down sides of the "OC-Mode" is that upon activation both Turbo and XFR will be disabled as well. Basically, this means that unless you are able to reach at least the default MSCF / XFR frequency on all cores, then you will essentially be losing single threaded performance compared to stock configuration.

Man muss den 1700(X) schon auf 4.1GHz übertakten, damit er die Single Thread Performance des 1800X erreicht. Zudem werden sämtliche Stromsparmaßnahmen deaktiviert, und der Verbrauch schnellt furchtbar hoch. Hauptgrund ist ein automatisches Offset, das die Spannungen um bis zu 0,15V absenkt und dieses durch OC abgeschaltet wird.

Speichertakt wichtiger als Timings:

The northbridge of Zeppelin is officially called as the data fabric (DF). The DF frequency is always linked to the operating frequency of the memory controller with a ratio of 1:2 (e.g. DDR4-2667 MEMCLK = 1333MHz DFICLK). This means that the memory speed will directly affect the data fabric performance as well. In some cases, it may appear that the performance of Zeppelin scales extremely well with the increased memory speed, however that is necessarily not the case.

In many of these cases the abnormally good scaling is caused by the higher data fabric clock (DFICLK) resulting from the higher memory speed, rather than the increased performance of the memory itself.

Höherer Speichertakt ist höherer Data Fabric Takt. Und deshalb steigt die Performance.

Danke @BlauX für den Link zu Anandtech!

Lange Rede, kurzer Sinn:

AMD Ryzen 1800X mit günstigem Mainboard und schnellen RAM kaufen und nicht übertakten.
 
Zuletzt bearbeitet: (Ryzen und Zeppelin vertauscht)
ich bin gespannt was die nächsten paar Wochen so bringen, vor allem wann nicht mehr alle paar Tage efi Updates kommen
Dann werde ich schauen ob es sich für meinen usecase lohnt meinen xeon 1230v3 zu ersetzen.
 
@ Raucherdackel!: DANKESCHÖN für dein Post!

Höchst interessant!
 
finde das ganze Thema extrem spannend:
  • teilweise weniger performance mit SMT,
  • Windows7 vs. 10,
  • Software, die nicht auf dem neusten Stand ist
  • BIOS-Ärger
  • Und Games, die wohl im Laufe der Zeit optimiert werden.


Alles in allem bin ich zufrieden mit der Performance, auch wenn es noch beim Launch ein paar Startschwierigkeiten gibt.
Jedoch muss man konstatieren, dass mit den ersten Erkenntnissen nach den Haupttests im Gegensatz zu Bulldozer die Hoffnung auf mehr Leistung in der Zukunft nicht unbegründet ist. Da schlummer definitv noch mehr Rohleistung.
 
Raucherdackel! schrieb:
Man muss den 1700(X) schon auf 4.1GHz übertakten, damit er die Single Thread Performance des 1800X erreicht. Zudem werden sämtliche Stromsparmaßnahmen deaktiviert, und der Verbrauch schnellt furchtbar hoch. Hauptgrund ist ein automatisches Offset, das die Spannungen um bis zu 0,15V absenkt und dieses durch OC abgeschaltet wird.

Dennoch ist der 1700 200€ günstiger als der 1800X, bis man den Preisunterschied durch den Stromverbrauch wieder reingeholt hat ist Ryzen längst ein nutzloses Stück Silizium.
Wenn ich mir ansehe was mein 2600k bei 4,4 GHz oder gar bei 5 GHz schluckt hab ich auch mit einem 1700 auf 3,8-4 GHz schon einen deutlich effizienteren Prozessor :D
Solange man sich selbst dran setzt und die perfekten Spannungen auslotet glaube ich nicht dass man spürbare Nachteile gegenüber dem 1800X hat, insbesondere da auch ein B350 Board den 1700 problemlos auf 3,8 GHz pushen kann wie man im Luxx bereits sieht.
 
@Raucherdackel: Danke für die Infos! Das ruft ja nach einer Ergänzung des hiesigen Artikels. ;-)
 
Raucherdackel! schrieb:
Man muss den 1700(X) schon auf 4.1GHz übertakten, damit er die Single Thread Performance des 1800X erreicht. Zudem werden sämtliche Stromsparmaßnahmen deaktiviert, und der Verbrauch schnellt furchtbar hoch. Hauptgrund ist ein automatisches Offset, das die Spannungen um bis zu 0,15V absenkt und dieses durch OC abgeschaltet wird.

Danke Dackel - Das würde dann auch den harten Stromverbrauch des Ryzen R7 1700 OC aus dem Hardforum ab 37x Multi erklären:

truespirit120-results.png
 
Zuletzt bearbeitet:
McTheRipper schrieb:
Oh Mann, da tun sich ja Abgründe auf...
Aber naja, "nur" ein Test.
@Raucherdackel: Das ist mal ne nette Info, danke dafür!

Das ist schon bedenklich, was hier CB macht. Dadurch werden Kaufentscheidungen vieler User negativ beeinflußt. Erinnert mich auch irgendwie an Intel, die mal vor ein paar Jahren über ihren Compiler absichtlich die AMD-CPUs ausbremsten. Sowas ist einfach nur noch unseriös!

http://techreport.com/news/8547/does-intel-compiler-cripple-amd-performance

Nachtrag: Ich respektiere die sehr gute Arbeit und Tests von CB. Nur solche Sachen wie oben genannt sollte man sich trotz aller Interessen verkneifen können. Weil das immer einen üblen Nachgeschmack hinterläßt und man sich noch später daran erinnert. Dadurch geht auch ein Teil des Vertrauens verloren.
 
Zuletzt bearbeitet:
also leute ich verstehe euch nicht wirklich
klar ist es unschön dass 1-2 benchmarks nicht der wahrheit entsprechen und auch das sollte geändert werden
aber das noch immer der alte spielebenchmarkparkours drin ist finde ich noch viel schlimmer
diesen mit grademal 6 spielen kann man einfach nur nur als schlecht bezeichnen
und man sieht ja auch wieviele sich ausschließlich den alten parkours anschauen

@ WommU
gestern war nicht sonntag und vorgestern abend auch nicht
 
Zuletzt bearbeitet:
Also ich unterstelle CB da keine böse Absicht. Kann doch gut möglich sein dass sie es nicht besser wussten.
 
Bitte vergesst nicht, dass eine Änderung der verwendeten Software nach sich zieht, dass alle Prozessoren neu getestet werden müssen, da die Vergleichbarkeit ansonsten nicht gegeben ist. Dazu hatte CB wahrscheinlich weder Zeit noch Geld, viele Testmuster sind ja immer nur für den Testzeitraum geliehen und stehen CB gar nicht mehr zur Verfügung. Das ist ja sicherlich auch der Grund, warum der neue Spielebenchmarkparcour nur eine geringe Anzahl an Prozessoren enthält.

Es ist natürlich wichtig, dass solche Umstände diskutiert werden und CB ist sicherlich auch dafür dankbar.
 
Irgendwie kann ich nicht immer noch nicht entscheiden... 1700x+x370 (etwas OC) Board oder 1800x+b350...
Vor allem scheint jeder eine andere Meinung zu RAM Speed und timings zu haben :/
 
motul300 schrieb:
@ WommU
gestern war nicht sonntag und vorgestern abend auch nicht

Schon mal dran gedacht, welcher Aufwand und Stress ein solcher Test macht? Inklusive Nachtarbeit. Da darf wohl mal ein freies Wochenende drin sein!!! Gestern war auch schönes Wetter. Soll er rund um die Uhrzeit permanent zur Verfügung stehen?
 
Zurück
Oben