"Geekerwan’s Q&A Live
Q. Is this year better than last year for upgrading devices?
A. I think so. This year, Qualcomm, MediaTek, and Apple have all made significant advancements. Compared to the previous generation, there have been major improvements, especially in terms of chipsets.
So, I believe it’s a good time. No matter which brand you prefer, if you’re considering an upgrade, it seems like a reasonable choice. After all, chipsets have a big impact on performance.
The most notable progress this generation has been in efficiency. And efficiency directly affects battery life. Additionally, this year, devices from Apple, Qualcomm, and MediaTek have all seen an increase in battery capacity.
Battery life has improved across the board compared to previous generations. Longer battery life and reduced heat generation—these are the factors that influence user experience, right? These are the biggest improvements.
Q. Are you planning to buy the Samsung S25 Ultra early next year?
A. I don’t think I will. I’m still deciding on what my next phone will be. I feel like I prefer smaller screen phones, so I’m considering going back to a compact device.
(Info: Current phone is Galaxy S23 Ultra in Space Black.)
Q. Over the past few years, it seems like whichever comes out first between Snapdragon or Dimensity tends to fall behind. Hmm.
A. First of all, last year, the Snapdragon 8 Gen 3 came out first but couldn’t outperform the Dimensity 9300, right?
This year, the Dimensity 9400 launched first, but the Snapdragon 8 Elite, which came later, ended up being stronger. You could say the 8 Elite is slightly better overall, especially in terms of efficiency.
The Dimensity 9200 generation was completely outclassed by the Snapdragon 8 Gen 2, wasn’t it? The 9200 couldn’t match the 8 Gen 2.
For this generation, the Dimensity 9400 is slightly behind the Snapdragon 8 Elite. When you look at aspects like battery life or gaming performance, the 8 Elite is a bit better overall.
That’s true; the Dimensity 9400 is more affordable in terms of its purchase price. However, it’s hard to say how this affects the final device price because that ultimately depends on how manufacturers decide to price their products
."
Also comments about energy efficiency curve:
"He also talked about the energy efficiency curve, which can be summarized as follows:
Let’s discuss low-frequency multi-core performance. Low-frequency scenarios are not very common in real-world usage. Such scenarios typically occur during idle states, standby modes, or ultra-light workloads.
The use of low frequencies differs from scenarios like energy efficiency curve testing, where all cores are run at 100% capacity. In low-load situations, only a single core or dual cores are active, operating at approximately 30-40% load. These light workload conditions deviate from the low-frequency energy efficiency curve.
Ultimately, battery life is the result of a combination of standby power consumption and medium-to-high load power consumption.
Standby power consumption is influenced by factors such as board optimization, the processor’s low-frequency state, and memory DVFS (Dynamic Voltage and Frequency Scaling) optimization. On the other hand, moments when we shop, use social media, watch video apps, load data, or scroll through content are associated with medium-to-high-frequency scenarios.
In these moments, medium or high frequencies are utilized, causing a sharp increase in power consumption. As a result, medium-to-high-frequency efficiency has a more significant impact on battery life in such scenarios.
This means that the low-frequency region of the energy efficiency curve is not directly applicable to real-world situations. In reality, during low-load states, only two or three cores are used, operating at about 30% load, rather than running all cores as is done during curve measurement.
When plotting the energy efficiency curve, running all cores is an unavoidable part of the testing process."
Source:
Jukanlosreve [X]