AMD Zen5 Ryzen™ 8000 exposed
In 2018, in a video officially uploaded by AMD, Mike Clark, the father of Zen architecture and chief designer, revealed for the first time that the team had begun zen5 research and development, when Zen+ had not yet officially debuted. The following year, David Suggs, one of the main designers of Zen 2 and an AMD enterprise academician, mentioned in his profile that he had joined the Zen 5 team to develop and design. In October of that year, AMD CTO Mark Papermaster confirmed that Zen 4 and Zen 5 were being independently developed by different teams. Since they do not interfere with each other, it seems reasonable that there will be a huge difference in the underlying architecture.
RedGamingTech gave the details of Zen 5 stealing in the latest revelations, including the first use of a large and small core hybrid architecture similar to Intel's 12th-generation Core, of which the Ryzen 9 series is a pure large core, and the corresponding Ryzen 8000 desktop CPU can reach up to 32 cores.
In terms of performance, Zen 4 to Zen 3's IPC increase is about 19%, while Zen 5's to Zen 4 is actually 30%, which is quite terrifying.
In addition, Zen 5 completely refactored the cache section, and the capacity of L1, L2, and L3 will increase significantly.
Although Zen 5 has not yet a clear time point, considering that Intel will iterate on the 13th generation of Core Raptor Lake in the second half of the year and follow up with the 14th generation of Core Meteor Lake next year, this is bound to accelerate the development process of AMD competitors, and it will be seen.
Vivo self-developed chip V1+ official announcement
In September last year, Vivo released its self-developed image chip V1, which can be matched with different main chips and screens, which can expand the ISP computing power and release the MAIN chip ISP load. This V1 chip has advantages in performance, latency, power consumption and other aspects of specific image processing tasks, compared with CPU, DSP and other chips, V1 processing specific tasks of efficiency has "exponential improvement", can also complete the parallel processing of data. Specific to the application scenario, in the night scene shooting, algorithms with larger demand for computing power such as denoising and interpolation can be applied. With the assistance of V1, the main chip can achieve 1080P 60 frames of night scene video noise reduction and interpolation.
With the blessing of V1, the main chip can run the MEMC denoising and interpolation of 4K 30FPS at low power when recording in low light, which assists and strengthens the image effect of the main chip in the night scene, and cooperates with the original noise reduction function of the main chip ISP to achieve secondary brightening and secondary noise reduction.
Today, vivo announced a new self-developed image chip V1+, this vivo self-developed image chip will be launched by vivo X80 series, the official said that this chip can do a "core" two uses, in addition to improving the image experience, there are additional surprises to be revealed at the press conference.
It is reported that the vivo X80 series equipped with V1+ chips will debut on April 25.
NVIDIA has begun testing RTX 4090 samples
According to the latest intelligence given by hardware whistleblower kopite7kimi, GeForce RTX 4090 graphics card has been confirmed, which uses the AD102 GPU core, matches 24Gbps high-speed video memory, and is currently in the testing stage.
Generally speaking, it takes about half a year from testing to final release on the market, according to which the RTX 40 series will appear at the end of the third quarter at the earliest.
According to previous data, the AD102 core is expected to hatch the RTX 4080, the difference is that it shields some computing units. It is widely estimated that the RTX 4090 will integrate up to 18,432 CUDA, a 71% increase over the RTX 3090 Ti. Correspondingly, the two-digit increase in performance is certain, depending on the size of the ten digits.
In fact, in terms of game performance, N card has always been the existence of the leader, for the RTX 40 series, other highlights should also be adapted to PCIe 5.0 16Pin power interface, output interface added DP2.0 and so on.
The bandwidth of DP2.0 is twice that of HDMI 2.1, and the 40Gbps of HDMI2.1 can transmit 4K 120 YUV444 10 bits at maximum, supporting HDR10 and Dolby Vision, and DP2.0 supporting 4K 240 YUV444 lossless 10-bit color, supporting HDR10 and Dolby Vision.
OnePlus Ace comes standard with longevity version 150W super flash charge
OnePlus has previously announced that it will hold a new product launch conference at 19:00 on April 21 to officially launch the OnePlus Ace series of mobile phones. In the past few days, the official has opened a full range of warm-ups, and announced a lot of important information about the new machine.
This morning, Li Jie, president of OnePlus China, warmed up the new OnePlus Ace again, and posted that the full version of the OnePlus Ace is standard and is equipped with a "longevity version of 150W super flash charge".
It is reported that this longevity version of the 150W super flash charge is not only charging very fast, but also the battery life can be 1 times longer than the industry standard, which can reach the standard of 1600 times.
More importantly, the OnePlus Ace will also come with a 150W fast charging head.
In addition to fast charging, another highlight of the OnePlus Ace is the performance specification, which will be the world's first MediaTek Tianji 8100-MAX chip, which is a chip deeply customized and deeply tuned by the OnePlus team and MediaTek.
On the basis of fully possessing all the characteristics of the Tianji 8100 "God U", the Tianji 8100-MAX has further improved its capabilities in the following three aspects:
1. The stability of the game frame rate has been greatly improved;
2. Ai computing power has been greatly improved;
3. The noise reduction ability of night scene video has been greatly improved.
Officials said that OnePlus has deeply integrated and redeveloped the game optimization technologies of HyperEngine 5.0 and HyperBoost, the underlying level of the chip, and the game frame rate is more stable, and the OnePlus Ace is greatly better than the industry in terms of frame rate stability.
For the first time, the OPPO K10 Pro series is equipped with an 80W Super Flash Charge
Today, OPPO announced that the upcoming K10 Pro will be the first mobile phone in the K series to support 80W wired flash charging while having a large 5000mAh battery.
According to the official introduction, OPPO K10 Pro will have an exclusive battery intelligent health engine, which ensures the stability and life of the battery under high-power fast charging, so that the 5000mAh large battery can be fully charged in only 31 minutes, realizing both fast charging and large battery.
In addition to fast charging, OPPO K10 Pro will be equipped with Qualcomm Snapdragon 888 flagship processor, which uses Samsung's 5nm process process by 1× 2.84GHz super core + 3× 2.4GHz large core + 4× 1.8GHz small core, gpu Adreno 660.
Photography, the K10 Pro is equipped with a Sony IMX766 sensor and supports OIS optical image stabilization, which is the first time that the K series has adopted this combination.
The OPPO K10 series will be released on April 24, when more details will be revealed.
AMD is about to release three new graphics cards
At present, there are many sources that AMD will release a new graphics card RX 6x50 XT series on May 10, including RX 6950 XT, RX 6750 XT, RX 6650 XT, so how do they coexist with the existing RX 6900 XT, RX 6700 XT, RX 6600 XT?
According to the latest reliable exposure, after the release of the RX 6x50 XT series, the corresponding RX 6x00 XT series will be discontinued and delisted at the same time. In other words, this time is not an upgrade and expansion of the army, but an iteration.
At present, AMD, NVIDIA graphics cards are gradually returning to the official original price, the supply is becoming more and more stable, RX 6600 XT, RX 6500 XT, etc. in some countries and regions have even been lower than the initial price.
Of course, this also means that it takes some time for the old and new cards to be replaced.
The only question is, will the RX 6x50 XT series stay at the same price, or will it take advantage of the opportunity to rise?
Judging from the current news, in addition to increasing the video memory frequency, the RX 6x50 XT series will also increase the core frequency and the power consumption of the entire card, bringing greater performance improvements.