Samsung EVO 860 1TB really slow in testing

Question

SSD

SSD（ソリッド・ステート・ドライブ）は、NANDベースのフラッシュ・メモリーを使用してデータを保存するストレージ・デバイスの一種です。

52回答すべてを表示

Adam Milton @adammilton

評価: 11

1

投稿日: 2020年4月27日

オプション

Samsung EVO 860 1TB really slow in testing

I bought a Samsung EVO 860 1tb SSD last fall to use in my MacBookPro (15”, late 2011), which is now dying, and now my iMac (27”, late 2012) needs a new HD, so I thought I would take this drive out of the MBP and use it in the iMac. I have the iFixit kit for iMac SSD installation, so the last step is confirming which SSD to use.

Before installing, I wanted to do some testing using Black Magic Disk Speed Test, here are the results:

iMac Fusion Drive (HDD+SSD):

Write: 150-200mb/s

Read: 250-400mb/s

Samsung 860 EVO:

Write: 30-50mb/s

Read: 220mb/s

So my question is, what’s going on here? It makes no sense at all that the fusion drive tests results are beating the pants off the EVO, which is giving worse results than just about any HD I can recall testing. I did test the speed via Thunderbolt 1 using Target Disk Mode, but also got similar results directly on the MBP that houses the EVO. The SSD is pretty full, would that have a drastic affect on the speed? Is this thing just bad?

Thanks in advance for any help!

この質問に回答する同じ問題があります

この質問は役に立ちましたか？

はいいいえ

スコア 0

4 件のコメント:

How are you connecting the drive via the USB3 connection or via a Thunderbolt drive case?

If you used a SATA to USB adapter or drive case you are bing limited to the USB connection.

Dan さんによる2020年4月27日

Hey Dan, I'm connecting the two machines with a Thunderbolt cable

Adam Milton さんによる2020年4月27日

The chaining you have done is too complex for any true measure of performance.

You have two systems PCH logic which has to be navigated when using Target Disk Mode (being the faster) and if you are using a networked connection via Thunderbolt you have the systems CPU's involved which then adds more delay into the measure.

Dan さんによる2020年4月27日

@danj That's a great point, I also did testing directly on the MBP, and the results typically start off better on the first test (W: 70mb/s, R: 330mb/s), but then after a few more tests, the results level off at W: 30mb/s, R: 300mb/s. The read is faster, but the write is the same, or worse when testing directly on the MBP.

The MBP has a bad GPU, so I can run the MBP for short periods of time before it eventually crashes.

Adam Milton さんによる2020年4月27日

コメントを追加

2件の回答

How are you connecting the drive via the USB3 connection or via a Thunderbolt drive case?

If you used a SATA to USB adapter or drive case you are bing limited to the USB connection. — Dan
Hey Dan, I'm connecting the two machines with a Thunderbolt cable — Adam Milton
The chaining you have done is too complex for any true measure of performance.

You have two systems PCH logic which has to be navigated when using Target Disk Mode (being the faster) and if you are using a networked connection via Thunderbolt you have the systems CPU's involved which then adds more delay into the measure. — Dan
@danj That's a great point, I also did testing directly on the MBP, and the results typically start off better on the first test (W: 70mb/s, R: 330mb/s), but then after a few more tests, the results level off at W: 30mb/s, R: 300mb/s. The read is faster, but the write is the same, or worse when testing directly on the MBP.

The MBP has a bad GPU, so I can run the MBP for short periods of time before it eventually crashes. — Adam Milton

Dan @danj 評価: 479.1千 Self employed contract worker 35 329 971 · Answer 1 · 2020-04-27T07:40:26-07:00

You will also see some issues with just swapping out your MacBook Pro HDD to SSD. As you didn’t tell us which model I can’t give you the needed steps to replace it and what will be needed.

I only use Samsung SSD's as they have a good track record and their performance is always at the top when you check the reviews.

I think your testing has hit its limits with your MacBook Pro as well with a bad GPU. Even still your number could be better if you replace the older HD SATA cable which has issues in this series.

Apple's original cable was only rated for SATA II (3.0 Gb/s) so when you use a high performance drive you can end up encountering performance issues as the cable just can't support the higher throughput. To fix this we install a better HD SATA cable which is designed to support SATA III (6.0 Gb/s) drives MacBook Pro 15" Unibody (Mid 2012) Hard Drive Cable.

We also need to help protect the cable from damage as Apple also encountered issues with the rough surface of the aluminum uppercase. Here we use a strip of electricians tape to protect the cable from wear

Your Hard Drive Cable Is A Ticking Time Bomb

The next thing we need to worry about is in the process of installing it we don’t damage the cable. The cable is made of thin foil wires which can be damaged when you fold the cable too far. A common issue I’ve found is people crease the cable with sharp crease when the cable needs to bend around a corner which fatigues the wires. Instead I use a old BIC pen ink straw as a forming brake to roll the cable into a nice smooth arc as the radius of the straw is as tight a curve you want in the cable.

Hey Dan, thanks for the question, the iMac is 27” Late 2012, and the MBP is 15” Late 2011, I have also edited the question for greater clarity! — Adam Milton

Yiu Wong @yiuwong 評価: 1 · Answer 2 · 2024-02-29T15:06:26-07:00

Yiu Wong @yiuwong

評価: 1

投稿日: 2024年2月29日

オプション

I had the same problem but I found a fix for it. In Samsung Magician select Performance Optimization and set it to Full Performance Mode. Once done set it back to Standard mode, read and write speeds should go back to normal.

このアンサーは役に立ちましたか？

はいいいえ

スコア 0

1件のコメント:

Sounds like you bought a used drive as I’ve never needed to alter the performance mode which by default is Full Performance.

Dan さんによる2024年2月29日

コメントを追加

SSD

Samsung EVO 860 1TB really slow in testing

2件の回答

Update (04/27/2020)

MacBook Pro 15" Unibody (Mid 2012) Hard Drive Cable

回答を追加する