RealTest on a Mac
RealTest’s documentation says it “is written in the C programming language using the native Windows API”, and that it “works fine on a cloud-based virtual machine, or a Mac running BootCamp or Parallels, provided that the VM emulates an x64 CPU”. If that setup works for you, it’s a fine answer.
Rearview is for when you’d rather test natively on your Mac, with an engine that is checked against a second engine on every release and tells you when a result is probably luck.
Running RealTest on a Mac today
- A Windows virtual machine. Boot Camp is available only on Intel Macs. On a Mac with Apple silicon, Parallels or VMware Fusion runs the Arm version of Windows, and Windows runs x64 programs on it through emulation. RealTest’s documentation adds that “API-level emulators such as Crossover or Wine” are not supported.
- Your data. If you use Norgate Data, its updater “will only work under Windows”, so it runs in the virtual machine too. Norgate’s FAQ adds that during an active subscription “you may export or extract data and process it on any other computer, operating system, server”, provided the use stays personal.
Moving a strategy to Rearview
A RealTest script and a Rearview script are organised much the same way, a section per rule, so most strategies move line by line:
| RealTest | Rearview |
|---|---|
Side: Long with EntrySetup: <condition> | enter long when <condition> |
ExitRule: <condition> | exit when <condition> |
EntryScore: <score> | rank by <score> (higher is preferred) |
Quantity: 10 with QtyType: Percent | size 10% of equity |
MaxPositions: 10 | max positions 10 |
EntryTime: NextOpen, ExitTime: NextOpen | fill next_open, the default |
ExitStop: FillPrice - 3 * atr20[1] | stop 3 x atr(20) |
A Data: section (sma200: MA(C, 200)) | bindings (sma200 sma(close, 200)) |
MA(C, 200), Highest(H, 30), Lowest(L, 20) | sma(close, 200), highest(high, 30), lowest(low, 20) |
x[1] | x[1] |
Several Strategy: blocks in one test | a portfolio file: two to eight strategies in fixed sleeves or sharing one pool of cash |
A Scan: section, its Filter formula and named formulas as columns | a scan script: filter <condition>, and each binding as a column |
This table is written from RealTest’s public documentation and hasn’t yet been checked by running the same strategies in both tools. If a line is wrong, tell us.
The language reference has every statement and function, and the examples are complete scripts to start from.
Getting your data across
If your data comes from Norgate, Rearview’s Norgate export script writes it in the import format from inside the virtual machine: delisted stocks, splits, dividends and, if your subscription includes past constituents, index membership by date. Save it from Data Library ▸ Import ▸ Norgate Export Script… or download it, run it where Norgate Data Updater runs, and import the folder it writes.
From other sources, export one CSV file per symbol and point Rearview at the folder with Data Library ▸ Import ▸ Per-Symbol Files…, or rvw convert ohlcv on the command line. The converter finds the columns by name and says what such files can’t tell it: whether prices are already adjusted, which stocks were delisted, and which were in an index on a past date. For those, use a source in Rearview’s import format.
What works differently
- Signals at the close, fills next session. Every rule is evaluated on a day’s close and orders fill in the next session, at the open by default. Same-day fills are possible with
fill same_close, and results that use them are marked approximate. - Prices as traded. Rearview stores the prices you actually traded at and works out split- and dividend-adjusted series itself.
closeis the day’s real close, averages run on split-adjusted prices, and percentage indicators on total-return prices. - Sizing and cash. A position is sized from the signal day’s equity. If cash runs short after a gap, the share count shrinks rather than the trade being skipped.
- Stops. A stop is live from the entry day, and a day that opens beyond it fills at the open.
- Ties. When candidates rank exactly the same, the one with the lower security id goes first.
- Is it luck? Every search counts the variations you tried, earlier runs and previews included, and reports a Deflated Sharpe Ratio and the Probability of Backtest Overfitting, with a plain verdict. A single run can be set against random entries at its own rate and stress-tested. A lockbox holds back the latest data until you choose to look.
Sources
Read on 23 October 2026: RealTest system requirements and Norgate Data Updater FAQ. Read on 3 November 2026: RealTest’s Scan section.
Product names belong to their owners. Rearview isn’t affiliated with or endorsed by them.