Sony XR晶片真的比較猛

MewTube
個人積分：5167分
文章編號：87746527

5167分

樓主

2023-06-28 5:02

身為鋼鐵索粉，這次來吹捧一下索尼XR晶片的威力
Rtings近期更新了評分標準 v1.11，
其中最大的改動就是增加了低訊源升頻的畫質表現，
以往Rtings把所有電視的升頻能力一律齊頭式的給8.0分，
所以過去評分標準對機種的升頻表現根本沒任何鑑別力。
Rtings於是使用480p訊源升頻成4K藉以觀察影像升頻後的清晰程度。

在Rtings當前測試的數十款電視中 (點我連結)，
很明顯Sony XR晶片是勝過各廠機種的表現，
XR晶片系列的電視(X95K、X90K、A95K等等)普遍拿下9.0分為全機種之冠，
低訊源升頻的還原度非常銳利自然。
其次為三星量子晶片和LG Alpha 9晶片、Sony 陽春X1晶片等等，
升頻能力大約落在7.0 - 8.0分之間。
Rtings給最低分的是TCL ，普遍落在3.0 - 5.5分之間。

Sony X95K
Sony XR晶片真的比較猛

Samsung QN95B
Sony XR晶片真的比較猛

LG C2

TCL R646 (約同等TCL C825規格)
Sony XR晶片真的比較猛

藉此可以看出索尼晶片粉並非全然迷信，
XR晶片是真的有兩把刷子 ^++^

當然價格上值不值那個價差就見仁見智...

2023-06-28 5:02 發佈

文章關鍵字 Sony XR 晶片

斯巴拉系大公

斯巴拉系大公
個人積分：2538分
文章編號：87746549

2538分

2樓

2023-06-28 5:48

A95K跟S90C

還有甚麼都不做只有正常倍線的U8H

此外這是RT給的原圖

其實照理來說RT的另一個測試，低質量內容平滑會比較重要，但屬我直言我看不懂他們那項測試在做甚麼，他們也沒解釋WwwwwW

羅掌櫃

羅掌櫃
個人積分：192分
文章編號：87747964

192分

3樓

2023-06-28 10:37

SONY的電視機就是好，畫質、功能等, 還是比其他廠牌來的令人喜愛

如果人品好, 可以用很久的話, 那就更好了

ThomasRhin

ThomasRhin
個人積分：1869分
文章編號：87748089

1869分

4樓

2023-06-28 10:51

用過LC C2與舊的sony LCD F系列電視(x1)，顏色校準與升頻sony勝出，LG給我的感覺就只贏在OLED面板天生特性。而幾年前索尼H系列的影像處理晶片傳出被整合進聯發科晶片(號稱客製晶片)，我都懷疑到底是不是變成僅韌體性質的影像處理功能而已，XR晶片現在是獨立晶片還是整合晶片，有人知道嗎？

心之死

LG的晶片沒有Sony好這是真的,但也不完全是天生面板優勢,最近買了一個中國貼牌的OLED才發現,中國貼牌才是真的用OLED面板天生優勢,軟體沒用,詳情可以看我的開箱文

2023-06-28 15:02

chanp

中國oled TV 主要依賴發哥晶片，發哥跌倒，客戶bug吃飽，s81是舊機種+舊晶片又不給更新，大部分情況能正常使用就不錯了，跟軟體無關

2023-06-28 18:32

chanp

chanp
個人積分：14752分
文章編號：87748193

14752分

5樓

2023-06-28 11:02

MewTube wrote:
身為鋼鐵索粉，這次來...(恕刪)

X95K

QN95B

LG C2

TCL R646

A95K

S90C

U8H
Upscaling: Sharpness Processing 缺

更新：
https://www.rtings.com/tv/tests/picture-quality/upscaling-sharpness-processing

UPSCALING: SHARPNESS PROCESSING

Full resolution 4k reference image

Our Upscaling: sharpness processing test determines how well a TV can upscale content with a lower resolution than the native resolution of the panel.

Upscaling mainly consists of two parts: the TV first has to scale the image to match the native resolution of the panel. For example, if a 4k TV is displaying a 1080p source, then it needs to increase the total resolution of the image by a factor of four. Once the TV has scaled the image, the TV's image processor goes over the image and applies extra filters, like noise reduction, antialiasing, and sharpness processing, to improve the image. The focus of this test is on the latter of those: the sharpness processing, as it's one of the most immediately noticeable aspects of picture quality.

Our test uses a 480p video with a non-square (anamorphic) pixel aspect ratio. Even though we're using a 480p signal, the results of this test are valid for any low-resolution or low-bitrate source, like DVDs or cable TV channels. As the quality of the signal increases, the sharpness processing of the TV is less important, though, as the TV doesn't have to work as hard to produce a nice image. Even going up to 1080p can make a huge difference, as most modern TVs do very little processing on 1080p signals like Blu-rays, so issues that are exposed with a 480p signal aren't as obvious.

The sharpness test pattern used to determine the optimal sharpness setting on each TV.

The first step in our upscaling test is to determine the ideal sharpness setting for each TV. We use a sharpness test image from the Spears and Munsil HD Benchmark 2nd Edition Bonus DVD and adjust the sharpness setting on the TV to find the maximum setting that introduces the least amount of artifacts or errors in the image. This ensures that straight lines remain straight and text is easy to read. We start by setting the sharpness setting to its maximum, introducing artifacts like digital noise and white padding around lines. We then lower the sharpness setting until those issues are no longer noticeable at a comfortable viewing distance.

Once we've determined the optimal sharpness setting for the TV, our next step is to evaluate the sharpness processing of the TV with our test video clip. We connect the TV to a PC and set the computer to send a 720x480 resolution with the output color format set to YCbCr, with 4:2:0 chroma subsampling, as this is the color format used by most DVD and low-resolution sources. We use MPC-HC to play our upscaling video on the TV in a loop and take multiple pictures.

The score is subjectively assigned based on the sharpness of the image and the quality of the upscaling. There are a few different points of comparison in the test clip that make it easy to see the differences in a TV's upscaling. The roof shingles should be clearly visible on a TV with good upscaling, whereas low-quality upscaling causes the shingles to blend together. Next, there's some text throughout the scene, including the names of the boats. Again, on a TV with good upscaling, the names should be easy to read. Finally, there's an overlay that closely mimics how well a TV can upscale hardcoded subtitles. Since each line is progressively smaller, it's easy to see where a TV starts to struggle so we can objectively score their performance. Most TVs begin to struggle at around the third line, where the "T" starts to be difficult to read and the "I" and "N" combine.

Cropped reference image.

Sony X90K - Fantastic 480p upscaling.

TCL 6 Series/R646 2021 QLED - Bad 480p upscaling.

It’s important to note that because the quality of 480p is so low, even top performers on this test won't produce an amazing picture. Above, compare a TV that delivers fantastic upscaling (center) with one that produces a softer image and isn’t as good (right). You can see some detail is lost in the softer picture, but also that neither looks great. Fine details on the TCL are hard to see, and text, especially the RTINGS text in the bottom-right of the image, isn't well defined, and edges appear soft.

...(恕刪)

Related Tests
Low-Quality Content Smoothing:
Whereas sharpness and upscaling look at a TV's ability to scale a lower-resolution image and display it cleanly, smoothing looks at how well a TV can process that image to smooth out rough areas of the content and reduce banding and macro-blocking caused by a low-bitrate signal. Whereas upscaling is mainly important for low-resolution content, smoothing is mainly important for low-bitrate content. Although many people believe that streaming services are generally high-quality, they usually aren't. Even if you have the highest tier available with your streaming platform, and even if it's still sending a 4k signal, almost all streaming services use compression algorithms to reduce the amount of bandwidth used, as bandwidth is expensive. This results in noticeable compression artifacts, such as black blocks in near-black scenes.

斯巴拉系大公

基本上照舊就行了，新版刪除了其他解析度的測試，然後在舊有的基礎上提高標準跟新的評分機制

2023-06-28 14:57

森元

森元
個人積分：934分
文章編號：87748422

934分

6樓

2023-06-28 11:29

我對rtings那個測評比較無感，應該說那個評測主要吃眼力及對畫質標準，每個人標準不一，像我朋友(是HD Club的高級發燒友)他測試Benq E65-730及65A80K，他說Benq在480P/720P升頻4K畫質比65A80K還好上一截，但我看了實機比較畫質完全不認為。

chanp

拿BenQ當例子基本上沒什麼說服力，其本身色準做的不錯，但???影像晶片???

2023-06-28 12:48

mieee77

Benq E50-730 看HD 4K片源並不差，畫質爛的是要升頻的非HD以上畫質片源。

2023-06-28 21:55

遺忘之村

遺忘之村
個人積分：319分
文章編號：87749589

319分

7樓

2023-06-28 13:53

可惜沒測國際牌 HCX Pro AI 跟東芝 Regza 晶片

chanp

基本上RT只測 USA有上市的機種…

2023-06-28 17:17

遺忘之村

我知道所以覺得有點可惜～

2023-06-28 18:12

八月二十

八月二十
個人積分：11775分
文章編號：87750056

11775分

8樓

2023-06-28 14:45

我的65X9000H真的很滿意，除了區域控光太少區域以外
不過我還是很想換75寸，有人要收嗎XD

oO喇叭智Oo

想了解一下，大大考慮賣多少

2023-06-28 17:54

HeeroYo

HeeroYo
個人積分：7544分
文章編號：87750120

7544分

9樓

2023-06-28 14:52

MewTube wrote:
身為鋼鐵索粉，這次來...(恕刪)

我永遠都記得燦坤sales一直要洗我買LG, 還一直攻擊SONY
但我最後說"我只想要SONY", 他就默默結帳了
SONY的好親眼看過就知道

CyouKei_ZERO

因為燦坤Sales賣LG TV台獎才會多，賣SONY 顯示器DP台獎很少......

2023-06-28 20:39

B000785

有時候是廠商駐點銷售穿燦坤的衣服，我也遇過穿燦坤衣服的猛推銷SONY電視，然後一直攻擊三星電視。後來我再去逛問電視時是另一個燦坤人員，態度就很中立，才知道一開始那個是SONY駐點

2023-06-30 8:23

eno03

eno03
個人積分：5275分
文章編號：87750142

5275分

10樓

2023-06-28 14:56

森元 wrote:
是HD Club的高級發燒友)他測試Benq E65-730及65A80K，他說Benq在480P/720P升頻4K畫質比65A80K還好上一截，但我看了實機比較畫質完全不認為。

精研所以前有些高手，後來有些是廠商或是店家，也算半個高手 ^++^

發燒友也很多類型，專攻兩聲道的，注重實體劇院的搭配，單玩耳機的都有，

通常這一類發燒用戶，對於電視跟顯示設備不見得鑽研很多，常眼見為憑
這兩台目前市價約四倍差，拿來比較時，對我來說已經很奇妙了

BenQ畫面較好的我目前認為只有S65-940，
其實它也就是上面的海信U8H改的

所有電視我看過升頻做最好的，個人覺得還是SONY
但不是每一台都好，也要到一定等級如X90之類
也如樓主所說，其中選擇的價差就見仁見智

通常我親友只看第四台或MOD，偶而看一點YT的
請他花4~6萬買SONY 65吋，他們會想很久

但兩萬元內的TCL，飛利浦，甚至海爾，雷鳥
他們看過，有些也就買來看了

較要求一點的則是抓了三萬元上下的國際

eno03

80吋當年還算不錯啦記得也要7萬，沒壞就懶得換吧，陳甯的YT偶而看一下，他有集電視水很深好像就在說SHARP好，SHARP日規新款OLED跟XLED評價不差，但台灣沒賣OLED

2023-06-28 17:50

xy.lin888

.

2023-07-02 18:13