A very nice monitor

Otis Grapsas

Carbonite Member
I recently got a monitor by Eizo, model 2333. It's a 23" led 1920x1080 monitor and it costs less than 500 euro with VAT.

Out of the box, in srgb preset, with a reduction of brightness to a reasonable level (about 100nits) and without calibration the performance is unreal. IMHO it performs a lot better than professional monitors in the 10,000-20,000 euro range including the Sony 23" 1920x1080 one. I would even say that the contrast ratio is better and the color error is a lot smaller. It doesn't have the features of a professional monitor but it is ideal for an editing station connected to a pc.

The online tests I found also verify the uncalibrated performance of this monitor, producing excellent gamma curve precision, dependable grays, true contrast ratios a lot better than 3,000:1, and color errors that are considered imperceptible by all known standards.

Eizo is probably the highest quality pc display manufacturer on the market and most of their monitors perform well, but I have never seen this level of performance in an SRGB preset before calibration, especially in an office monitor.
 
Samsung does general business/gaming displays, not in the same league with Eizo. This one is good compared to the competition, but not to the Eizo 2333. Here is a P2370 review. Different from your friends model probably but representative of a good business display.

http://www.prad.de/en/monitore/review/2009/review-eizo-ev2333wh-bk-part10.html

http://www.prad.de/en/monitore/review/2009/review-samsung-p2370-part10.html

The 2370 black point is more than 5 times brigher than the Eizo, so no deep blacks on the Samsung.

The 2370 color in sRGB standard has large errors in blue and red, DeltaE of more than 4.5. Even its mid greys are about 3.0 off. Not bad for a general monitor but the EIZO deltaE is only 0.9 and 1.4 for these colors and on mid greys it's about 0.5. Lower than 2.0 is excellent.

Also, the 2370 panel is the cheaper TN technology, not PVA. PVA has totally clean shadows, and true 24bit color, not sure what type of TN this is, but most (all I suspect) TNs are 18bit (6bit with dithering, not 8bit per color). If you see noise/activity in the shadows of a static image via DVI, the panel is 18bit. Generally PVA have smooth gradations compared to a TN panel. TN has a narrow vewing angle with lots of color shift and constrast changes from the side/top compared to PVA. The only advantage of TN is faster response, but that's not critical for color correction. PVAs are fast enough today.

All monitors are relatively good with calibration but they loose color in the process. Also, a profile is not active on every application so a good sRGB out of the box is very useful.

If you have good color precision on the sRGB preset and the calibration results in a relatively linear curve, that means the monitor gamma was precise, so you can trust that the uncalibrated monitor has excellent precision in all colors, including skin tones etc, not only the test colors (primaries and greys).
 
Last edited:
According to some reviews, the P2370 is better than the XL2370 in color precision. They are both TN. The XL2370 has LED backlight.

EDIT: Put simply, the XL2370 is an ultra bright consumer display that has problems with deep black tones, dithered 6bit color, average contrast, bad viewing angles, and a cheap panel. But don't tell your friend:)

Here is a comparison of a typical PVA panel and the XL2370 in the shadows:

http://pureoverclock.com/review.php?id=841&page=7

Blind in the shadows and innacurate gamma due to low 6bit precision.

And viewing angles of PVA (eizo) vs TN (2370) which affect brightness judgment and shadow/bright levels when correcting:

http://www.prad.de/images/monitore/eizo_ev2333wh/EV2333WH_blickwinkel.jpg

http://www.prad.de/images/monitore/samsung_p2370/samsung_p2370_blickwinkel_thumbnail.jpg

Drastically better, far higher contrast, high color precision, no color inversion etc. Viewing angle is very important in large displays, because even at the perfect viewing position you are viewing different parts of the screen at very different angles. Top to bottom angle problems can make shadow corrections very hard.
 
Last edited:
Btw, on older monitor specsheets, 18bit panels with dithering and frame rate control (FRC) had 16.2 million colors instead of 16.7 million colors in their color specs. So, the engineers could tell which monitors were capable of only 262K colors before dithering and which had noise/dithering issues. But most manufacturers made everything 16.7 million when this detail became widely known because 6bit color sounded very bad in marketing.
 
http://www.xbitlabs.com/articles/monitors/display/samsung-touchofcolor2_12.html#sect0

An XL2370 review. Check out how wrong the blue gamma is.

http://www.xbitlabs.com/images/monitors/samsung-touchofcolor2/6g2.png

A skin tone correction on this monitor will make skin tones (and everything else) very cold/blue because one will be compensating for this blue problem. The same problem you get when you mix music on a monitor with booming bass. You end up with thin bass on the average system (or any good studio monitor).

The Eizo correction curve (the reverse is the original response):

http://www.prad.de/images/monitore/eizo_ev2333wh/EV2333WH_farbk.jpg

It needed virtually no change.

The Samsung P2370:

http://www.prad.de/images/monitore/samsung_p2370/samsung_p2370_srgb_kurve_thumbnail.jpg

A more expensive Samsung, the 2380 which uses cPVA:

http://www.prad.de/images/monitore/samsung_f2380/samsung-f2380-korrekturkurve.jpg

Artificially dark/compressed shadows.

By now, it should be quite obvious how rare the performance of the Eizo 2333 is.
 
A useful monitor test:

http://www.lagom.nl/lcd-test/

The images for offline viewing:

http://www.lagom.nl/lcd-test/lagom-lcd-test.zip

Check these in every app you use for correcting color. You will see monitor problems and you might even discover application processing bugs.

The four critical tests are:

1) gamma-test-2.png Gamma precision for grey and each primary. Everything should blend in at 2.2. Otherwise, you might see exaggerated noise, images with wrong contrast, and large errors on skin tones because the midrange color mix will be off.

2) blacktest.png Shadow detail. This should show step 1 with marginal difference from the background and steps 2 and up clearly. Otherwise you will see crushed blacks where there are none or exaggerated noise and quantisation in compressed codecs.

3) whitetest.png Highlight detail. This should show step 254 with marginal difference from the background and steps 253 and lower clearly. Otherwise you might think something is clipped when it's not.

4) gradient-h.png Color quantisation. This should show a smooth gradient. Otherwise you might see color quantisation where there is none (skies, walls etc).

The problem with a bad reference is that we always try to compensate and this creates something that looks different on other monitors. The average of all existing monitors is sRGB, their response revolves around the sRGB standard, so that's the reference we need to approach in our monitor.

As a sidenote, most (if not all) expensive monitors that have very wide gamut, such as those used in prepress and photography, are usually incapable of achieving good precision on the smaller sRGB. There are not designed for sRGB and the wide gamut imposes technical limitations in sRGB reproduction.
 
Thank you, sir. That all explains a lot about what I've been seeing and not seeing on various LCD monitors over the past 5 years.

.
 
ill have to pick one up; i have an ancient CRT unit at the moment that is starting to make odd noises on startup and takes a while to warm up.
 
@Otis: HI! I read the 2333's review on Prad.de and I'm pretty interested in this monitor! Do you think is good for color grading due to its high contrast?
 
Last edited:
Back
Top