Some hours ago, during a discussion on Twitter, it came to light that a lot of people don’t understand the importance of a complete and clear review methodology.
Even geeks, engineers, journalists seem to be unable to see how bad an incomplete testing method could be. We know that in scientific fields a rigorous methodology is needed. According to scientific studies, a test shall be capable of being replicated: “Reliability is the degree of consistency of a measure. A test will be reliable when it gives the same repeated result under the same conditions” (Source).
To do this, we have to know exactly how the experiment was done. About a hardware review, we have to know every single hardware and software component was used, and which version of it. Thanks to these informations, if we want, we could be able to replicate the review tests and we could be able to confirm the achieved results.
So, during that Twitter discussion, someone questioned my results of this review, where I tested the power consumption of the Intel Core i9-9900K @ stock setting. As you can see, I meticulously described the configuration and the methodology I used. Also, I used default settings for everything: “out-of-the-box settings”. I did that in order to see how the PC of "average Joe" perform. We all know that average Joe purchases the pc, puts the plug and pushes the power button. Stop.
People complained that I had to change BIOS options in order to use a lower Power Limit 2, according to Intel specifications. The problem is that Intel itself allowed Asus, Gigabyte, MSI and other manufacturers to use that setting on high end main boards: PL2 @ Auto (About 210W). According to Intel, it’s not overclock: “The values on the data sheet for 1.52 V and 138 A are the maximum values that can be used on the Intel® processor” (Source). So, main boards manufacturers didn’t violate the specification. They just use the max PL2 value allowed.
But this is not the problem.
The problem concerns the methodology of the tests. These people, if they think my results are wrong, could remake the same tests with the same hardware and the same software. Despite that, they attacked B&C saying that other sites are more reliable. It could be. We absolutely don’t think B&C is the best site. However, I think B&C is the site with the best and more clear methodology tables. Here there are some examples.
Some Intel Core i9-9900 reviews (I can't list every 9900K review made)
As you can see, B&C tests can be replicated without problems. We don’t fear about that. On the other hand, we could have a problem if we want to replicate the tests of other sites.
In conclusion, we think that review sites have to improve their methodology, in order to create more reliable reviews. Currently, the reviews are far far away from being considered “scientific”.
Please note: It's not an attack to the other sites. B&C is a small site, attended in our spare time, so even we want to read some "scientific" reviews. We are end users in first place, not journalists. ;)
Article by Federico Barutto
The way Vega GPUs work is very particular, thanks also to a bad description of a fundamental WattMan setting. “HBM2 voltage” isn’t actually the voltage of the HBM2 chips (fixed at 1.2V on RX Vega 56s, and 1.35V on RX Vega 64s and Frontier Edition), but is more like a GPU voltage/frequency threshold.
By lowering this “HBM2 voltage” (Vbase, shortly) to 850mV you have the opportunity to undervolt nearly as much as you want the GPU, heavily reducing power consumption without losing too much performance if the card is water cooled, or actually enhancing them if the card is air-cooled.
If the power limit (PL) is kept at 0%, every GPU voltage (Vcore) over 1130mV is useless, since the card reaches its 220W (=~295W to the PCIe connectors) PL. On a watercooled card, GPU frequency is around 1580MHz.
By lowering Vcore to 950mV, GPU frequency lowers a bit to ~1530-1550MHz, but power consumption lowers dramatically to ~145W during rendering and around 115-165W during gaming.
And by lowering Vcore to 800MHz (lowest permitted voltage in WattMan/OverdriveNTool) frequency lowers to 1420MHz and consumption lowers around 125W during rendering and around 85-145W during gaming!
By doing that there’s a (non-) problem: the lower is the Vcore the higher is the lowest HBM frequency which makes possible to set the custom Vcore and Vbase. For example, with Vcore at 950mV and Vbase at 850mV this frequency is 965MHz, and with Vcore 800 and Vbase 850 it becomes 1000MHz… This isn’t bad, since Vega GPUs (as nearly every recent Radeon) love memory bandwidth. Raising the bandwidth from 484GB/s (945MHz) to 563GB/s (1100MHz) performance increases without any power increase. Think about that: rendered frames by FC5 go from 4750 to 4900 (+3%) by simply overclocking the HBM2 to 1100MHz, and LuxMark scores go up from ~31000 to ~33600 (+8.5%)!
People, sometimes, wonder what compilers are used in game development, since the game developers always try to squeeze every inch of performance out their engines. In reality, games don't really need any kind of special super-optimizing compiler. Games developers use different compilers, such as LLVM, Microsoft Visual Studio, ICC, etc.On the other hand, every compiler take advantage of specific CPU features in different ways. This doesn’t mean that a specific compiler is biased. This means that a compiler has some advantage over a rival compiler in some field.With this article, we want to see the behavior of Zen+ and Coffee Lake uArchs in old and recent video games, with and without uArch optimizations.
All the configurations are Spectre and Meltdown protected.
The tests follow these rules:
Last year we talked about the current commercial strategies of Intel, NVIDIA and AMD, but now we want to analyze just the strategies of AMD.
Former CEO Rory Read has paved the way, and now the CEO Lisa Su is doing a great job making AMD great again (Trump citation needed). But how?
Firstly, we have to quote Jerry Sanders, first AMD CEO: “People first, products and profit will follow”. Thanks to this idea, AMD has grown to the point of becoming the main Intel competitor. Despite the limited resources, AMD has been able to create some little gem, like K7, Hyper Transport, and K8.
Rory Read, as CEO, has taken up Sanders’ idea, creating a team of senior "pissed-off" engineers (Zen/K12 team), and calling back some never forget AMDers: Raja “The Oracle” Koduri, Terry “Catalystmaker” Makedon, and so on. Do you remember Mr. Wolf in Pulp Fiction? "I'm Rory Read. I solve problems".
Now, Lisa Su must insure that these efforts will be successful.
Pagina 1 di 3
Le recensioni e gli articoli che pubblicheremo sul nostro portale (77 messaggi)Ultimo messaggio di: Scaveon (16/10/2019 15:11)
[Official Topic] AMD "Zen" (2579 messaggi)Ultimo messaggio di: mafferri (14/10/2019 23:08)
Problema NAS - aMule (22 messaggi)Ultimo messaggio di: Bivvoz (13/10/2019 11:27)
[Official Thread] AMD GPU Navi RX5000 (704 messaggi)Ultimo messaggio di: Masciale (12/10/2019 00:41)
Che state ascoltando in questo momento? (35 messaggi)Ultimo messaggio di: ragen-fio (07/10/2019 16:23)
Buongiorno e buonasera (5385 messaggi)Ultimo messaggio di: Fottemberg (01/10/2019 15:33)
Ram per nas (6 messaggi)Ultimo messaggio di: mafferri (25/09/2019 21:18)
Una barzelletta allunga la vita (9 messaggi)Ultimo messaggio di: Scaveon (23/09/2019 14:57)
[Guida] all'overclock di AMD Ryzen. Il thread per spiegare ai novizi come overclockare le CPU Ryzen, per raccogliere pareri e per scambiarci suggerimenti.
[Official Topic] AMD APU Kabini, Temash, Kaveri e... Carrizo. Discutiamo le ultime novità sulle APU AMD basate su microarchitettura Bulldozer e Jaguar.
[Waiting for] AMD GPU Polaris & Vega (ex-Arctic Islands). Discutiamo le ultime novità sulle GPU Polaris di AMD, senza dimenticarci di Vega!
[Official Topic] Nvidia Pascal GPU. Discutiamo le ultime novità sulle GPU Pascal di NVIDIA e su quanto ci aspetta in futuro!
[Official Topic] AMD "Zen". Discutiamo le ultime novità sulla microarchitettura x86 Zen di AMD e le CPU/APU in arrivo.
Creative Sound Blaster e Windows 10/8.1/8/7/Vista. Dedicato a tutti i possessori delle schede audio di Creative Labs!
B&C e la questone "degli articoli prezzolati". Perché B&C è un portale diverso?